Mississippi State University Mississippi State University

Scholars Junction Scholars Junction

Theses and Dissertations Theses and Dissertations

12-14-2018

Extraction Kinetics, Physicochemical and Functional Properties of Extraction Kinetics, Physicochemical and Functional Properties of

Proteins Isolated from Channel Catfish (Ictalurus Punctatus) By-Proteins Isolated from Channel Catfish (Ictalurus Punctatus) By-

Products Products

Yuqing Tan

Follow this and additional works at: https://scholarsjunction.msstate.edu/td

Recommended Citation Recommended Citation Tan, Yuqing, "Extraction Kinetics, Physicochemical and Functional Properties of Proteins Isolated from Channel Catfish (Ictalurus Punctatus) By-Products" (2018). Theses and Dissertations. 2195. https://scholarsjunction.msstate.edu/td/2195

This Dissertation - Open Access is brought to you for free and open access by the Theses and Dissertations at Scholars Junction. It has been accepted for inclusion in Theses and Dissertations by an authorized administrator of Scholars Junction. For more information, please contact scholcomm@msstate.libanswers.com.

Template C v3.0 (beta): Created by J. Nail 06/2015

Extraction kinetics, physicochemical and functional properties of proteins isolated from

Channel catfish (Ictalurus punctatus) by-products

By

TITLE PAGE

Yuqing Tan

A Dissertation

Submitted to the Faculty of

Mississippi State University

in Partial Fulfillment of the Requirements

for the Degree of Doctor of Philosophy

in Food Science and Technology

in the Department of Food Science, Nutrition and Health Promotion

Mississippi State, Mississippi

December 2018

Copyright by

COPYRIGHT PAGE

Yuqing Tan

2018

Extraction kinetics, physicochemical and functional properties of proteins isolated from

Channel catfish (Ictalurus punctatus) by-products

By

APPROVAL PAGE

Yuqing Tan

Approved:

____________________________________

Sam K. C. Chang

(Major Professor)

____________________________________

Peter J. Bechtel

(Committee Member)

____________________________________

Ramakrishna Nannapaneni

(Committee Member)

____________________________________

Jiaxu Li

(Committee Member)

____________________________________

Marion W. Evans, Jr.

(Graduate Coordinator)

____________________________________

George M. Hopper

Dean

College of Agriculture and Life Sciences

Name: Yuqing Tan

ABSTRACT

Date of Degree: December 14, 2018

Institution: Mississippi State University

Major Field: Food Science and Technology

Select Appropriate Title: Add the name(s) of the person(s) heading your committee.

Title of Study: Extraction kinetics, physicochemical and functional properties of proteins

isolated from Channel catfish (Ictalurus punctatus) by-products

Pages in Study 205

Candidate for Degree of Doctor of Philosophy

Channel catfish raising is an important aquaculture in the United States.

Mississippi ranks No.1 in catfish production for continuous 4 years. However, catfish

fillet processing generates huge amount of by-products, including heads, bone frames,

visceras and skins, which contain significant amount of proteins. Removing and

utilization of proteins from the by-product would add value to the catfish industry.

Collagen was extracted from catfish skin by: (1) acid extraction; (2)

homogenization-aided; and (3) pepsin-aided extraction methods. Kinetic analysis of the

extraction was performed. SDS-PAGE was performed to analysis the composition of

proteins in collagens extracted under different conditions. Collages extracted by three

methods was characterized by solubility test, zeta potential, circular dichroism and gel

strength. Protein recovery rate from minced skin extracted with pH 2.4 HCl containing

23.6 KU/g pepsin was the highest (64.19%).

Papain, ficin, bromelain, neutrase, alcalase, protamex, novo-pro D and

thermolysin were used for the hydrolysis of the mixture of heads and frames. Proteolytic

activities of these proteases were examined using AzocollTM as a standard substrate.

Degree of hydrolysis (DH) of hydrolysates and enzyme reaction kinetics were

investigated. Results indicated that thermolysin had the highest activity (82.9×107

AzU/g) at 50 °C when using AzocollTM as the substrate. Ficin (80 AzU/g) was the most

efficient in hydrolyzing the proteins in the ground catfish by-product (DH reaching

71.88%) in 120 min at 30°C among all the enzymes. Thermolysin could be used for

industries to hydrolyze protein by-products in terms of hydrolysis efficiency and

economy. The hydrolysis curves fit the Peleg model very well, all with R2 higher than

0.91.

Myofibrillar proteins were extracted from the mixture of heads and frames with

different pH conditions, and made into protein gels. Transglutaminase (TGase) was

incorporated to improve the gel structure. Solubility and secondary structures of extracted

myofibrillar proteins were studied. Gelling properties of the protein sol were studied by

dynamic rheological measurement. Physicochemical, textural and thermal properties of

protein gels treated with TGase were investigated. Protein pattern changes of TGase

treated protein gel was studied by SDS-PAGE. Results indicated that alpha-helicity of

myofibrillar proteins decreased with extraction pH over 9. Storage modulus (G’) of

protein sol decreased as the increase of extraction pH. Rheological measurement of

TGase treated protein sol showed that excessive TGase could weaken the gel structure.

This study provides systematic information for the catfish fillet processing industry to

utilize the by-products.

Keywords: Channel catfish, by-products, collagen, extraction kinetics, hydrolysis,

myofibrillar proteins, secondary structure, protein gel.

ii

DEDICATION

I would like to dedicate this dissertation to my parent, Baosheng Tan and Suhua

Xie; my parents in-law, Dr. Qinghai Meng and Lijun Zhao for their endless love, support,

and understanding that make my graduate work a reality. And gratitude also to my

special lab-mate: my husband, Dr. Shi Meng, who at various occasions stood by me,

supporting and helped me to proceed forward.

iii

ACKNOWLEDGEMENTS

I would like to express deepest gratitude to my major advisor, Dr. Sam K.C.

Chang, a world-renowned professor, for accepting me into his research group. His deep

involvement and invaluable comments throughout my graduate study are greatly

appreciated, and his hard-working spirit and serious attitude will always be an excellent

role model for my work and study later on. Special gratitude also goes to Dr. Chang’s

wife, Mrs. Chiu-Ie Chang for her caring during my graduate study.

I also like to express sincere appreciation to my other graduate committee

members, Dr. Peter Bechtel, Dr. Jiaxu Li, and Dr. Ramakrishna Nannapaneni for their

helpful suggestions, equipment support, help on ordering chemicals, and the interest they

have maintained in this research.

I would also like to express appreciation to Dr. Rivers Orion at Institute for

Imaging and Analytical Technologies, Mississippi State University, for his help and

training to conduct scanning electron microscope experiment.

I would like to recognize the Mississippi Agricultural and Forestry Experiment

Station, USDA-ARS SCA No. 58-402-2729, USDA-NIFA AFRI 20186702228050 and

USDA-NIFA AFRI 20176701426264 contributed funding for providing my tuition, and

stipend in addition to research chemicals/supplies.

I appreciated Mr. Parker Lee (Country Select Catfish Co., Isola, MS) for his help

and the donation of catfish and by-products.

iv

I also wish to extend my deep thanks to our lab members Dr. Yan Zhang, Mr. Yin

Zhang, Dr. Yuwei Wu and Mr. Ruiqi Chen for their help and support.

My gratitude goes to all my good friends in Starkville, Dr. Lei Cao, Dr. Hsin-Yi

Lu, Dr. Xue Zhang, Dr. Li Zhang, Mr. Yin Zhang, Dr. Yan Zhang, Dr. Yuwei Wu and

Mr. Wenjie Shao. They have never hesitated to help, encourage, and support me. I am so

grateful to have such sincere friendship during my graduate study.

My gratitude also goes to my close friends in China, Qian Liao, Lan Yu, and

Wencheng Zeng, for their encouragement and unconditional support.

Finally, I would also like to express my appreciation to the department staff, Ms.

Donna Bland, Ms. Mary Mandol, Ms. Jessica Rahim, and Ms. Kerri Kelly for their kind

help and support.

v

TABLE OF CONTENTS

DEDICATION ................................................................................................................ii

ACKNOWLEDGEMENTS ........................................................................................... iii

LIST OF TABLES .......................................................................................................... x

LIST OF FIGURES ........................................................................................................ xi

CHAPTER

I. INTRODUCTION ............................................................................................. 1

II. LITERATURE REVIEW .................................................................................. 6

2.1 Aquaculture in the United State .............................................................. 6 2.2 Overview of catfish production .............................................................. 6

2.3 Overview of sales and consumption in US. ............................................ 7 2.3.1 Catfish sales in US ........................................................................... 7

2.3.2 Catfish consumption in the US ......................................................... 8 2.4 Catfish fillet processing.......................................................................... 9

2.4.1 Removal of skin, heads, frames and gut ............................................ 9 2.4.2 By-products from catfish fillet processing ........................................ 9

2.5 Overview of catfish by-products .......................................................... 11 2.6 Fish skin .............................................................................................. 12

2.6.1 Extraction and characterization of collagen .................................... 13 2.6.1.1 Extraction method .................................................................... 13

2.6.1.2 Secondary structure of collagen ................................................ 14 2.6.1.3 3-D structure of collagen .......................................................... 17

2.6.1.4 Peptide mapping and amino acid composition of collagen ........ 18 2.6.1.5 Fourier transform infrared spectrum of collagen ....................... 19

2.6.1.6 Solubility of extracted collagen ................................................ 21 2.6.1.7 Gel strength of collagen-induced gel ......................................... 22

2.7 Heads and frames ................................................................................. 23 2.7.1 Overview ....................................................................................... 23

2.7.2 Hydrolysis of by-products .............................................................. 23 2.7.3 Proteolytic activity determination ................................................... 25

2.7.4 Emulsion properties of by-product hydrolysates ............................. 26 2.7.5 Antioxidant activity of fish protein hydrolysates............................. 26

vi

2.7.6 Angiotensin-converting enzyme (ACE) inhibition activity of

fish protein hydrolysates................................................................. 27

2.7.7 Dipeptidyl peptidase IV (DPP- IV) inhibition activity of fish

protein hydrolysates ....................................................................... 27

2.8 Overview of viscera ............................................................................. 28 2.9 Surimi .................................................................................................. 29

2.9.1 Historical review of surimi ............................................................. 29 2.9.2 Manufacture of surimi .................................................................... 29

2.9.3 Manufacture of fish ball ................................................................. 30 2.9.4 Protein component of surimi (washed fish meat mince) .................. 31

2.9.5 Myofibrillar proteins extraction method ......................................... 33 2.9.5.1 Alkaline extraction ................................................................... 33

2.9.5.2 Salt extraction .......................................................................... 33 2.9.5.3 Acid extraction ......................................................................... 34

2.9.5.4 Extraction temperature ............................................................. 34 2.9.6 Gelation chemistry ......................................................................... 35

2.9.7 Microstructure of surimi and surimi like products .......................... 36

III. ISOLATION AND CHARACTERIZATION OF COLLAGEN

EXTRACTED FROM CHANNEL CATFISH (ICTALURUS

PUNCTATUS) SKIN ...................................................................................... 39

3.1 Abstract ............................................................................................... 39 3.2 Introduction ......................................................................................... 40

3.3 Materials and methods ......................................................................... 41 3.3.1 Materials ........................................................................................ 41

3.3.2 Chemicals ...................................................................................... 41 3.3.3 Proximate analysis ......................................................................... 42

3.3.4 Extraction of collagen with acids (ASC) ......................................... 42 3.3.5 Extraction of collagen with homogenization-aided (HSC)

method ........................................................................................... 43 3.3.6 Extraction of collagen with pepsin and homogenization aided

(PHSC) method .............................................................................. 43 3.3.7 Kinetic analysis of collagen extraction ........................................... 44

3.3.8 Sodium-dodecyl-sulfate polyacrylamide gel electrophoresis

(SDS-PAGE) .................................................................................. 44

3.3.9 Effect of pH on solubility ............................................................... 45 3.3.10 Effect of NaCl on solubility ............................................................ 45

3.3.11 Zeta (ζ) potential ............................................................................ 45 3.3.12 Circular dichroism (CD) ................................................................. 46

3.3.13 Differential scanning calorimetry (DSC) ........................................ 46 3.3.14 Gel strength .................................................................................... 47

3.3.15 Statistical analysis .......................................................................... 47 3.4 Results and discussion ......................................................................... 47

3.4.1 Proximate analysis of catfish skin ................................................... 47 3.4.2 Collagen extracted with acids and proximate analysis..................... 48

vii

3.4.3 Collagen extraction-aided by homogenization with or without

pepsin ............................................................................................ 49

3.4.4 Kinetic analysis of collagen extraction ........................................... 53 3.4.5 SDS-PAGE of collagen extracted with different conditions ............ 60

3.4.6 Effect of pH and NaCl on collagen solubility ................................. 64 3.4.6.1 Effect of pH on collagen solubility ........................................... 64

3.4.6.2 Effect of NaCl on collagen solubility ........................................ 64 3.4.7 Zeta (ζ) potential ............................................................................ 66

3.4.8 Circular dichroism (CD) ................................................................. 68 3.4.9 Differential scanning calorimetry (DSC) ........................................ 68

3.4.10 Gel strength .................................................................................... 73 3.5 Conclusion ........................................................................................... 73

3.6 Research diagram in proposal .............................................................. 75

IV. COMPARING THE KINETIC OF THE HYDROLYSIS OF BY-

PRODUCT FROM CHANNEL CATFISH (ICTALURUS

PUNCTATUS) FILLET PROCESSING BY EIGHT PROTEASES ................ 76

4.1 Abstract ............................................................................................... 76 4.2 Introduction ......................................................................................... 77

4.3 Material and methods ........................................................................... 78 4.3.1 Materials ........................................................................................ 78

4.3.2 Chemicals ...................................................................................... 78 4.3.3 Proteolytic activity determination by using Azocoll as

substrate ......................................................................................... 79 4.3.4 Proteolytic activity determination by using Azocasein as

substrate ......................................................................................... 80 4.3.5 Enzymatic hydrolysis ..................................................................... 80

4.3.6 Proximate analysis ......................................................................... 81 4.3.7 Determination of the degree of hydrolysis (DH) using

trinitrobenzenesulfonic acid (TNBS) method .................................. 81 4.3.8 Kinetic analysis of hydrolysis processing ....................................... 82

4.3.9 Calculated price of proteases activity and cost to reach 15 and

20% DH ......................................................................................... 83

4.3.10 Sodium-dodecyl-sulfate gel electrophoresis (SDS-PAGE) .............. 83 4.3.11 Emulsion capacities ........................................................................ 83

4.3.12 Foaming capacities ......................................................................... 84 4.3.13 Oxygen radical absorbing capacity (ORAC) of selected

hydrolysates ................................................................................... 85 4.3.14 Radical DPPH scavenging activity of selected hydrolysates ........... 86

4.3.15 Angiotensin-converting enzyme (ACE) inhibition assay ................. 86 4.3.16 DPP-IV inhibition assay ................................................................. 87

4.3.17 Statistical analysis .......................................................................... 88 4.4 Results and discussion ......................................................................... 88

4.4.1 Proximate analysis ......................................................................... 88

viii

4.4.2 Determination of proteolytic activity by using Azocoll and

Azocasein as substrate .................................................................... 89

4.4.3 Degree of hydrolysis of catfish protein hydrolysates ....................... 94 4.4.4 Kinetic analysis of hydrolysis processing ..................................... 112

4.4.5 Predicted cost of protease for approaching certain DH .................. 117 4.4.6 Sodium-dodecyl-sulfate gel electrophoresis (SDS-PAGE) ............ 118

4.4.7 Emulsion capacity and stability .................................................... 132 4.4.8 Foaming capacity and stability ..................................................... 136

4.4.9 Antioxidant activity of selected hydrolysates ................................ 137 4.4.10 ACE inhibition assay .................................................................... 145

4.4.11 DPP-IV inhibition assay ............................................................... 146 4.4.12 Protein recovery rate of hydrolysates with good

functionalities............................................................................... 150 4.5 Conclusion ......................................................................................... 151

4.6 Research diagram in proposal ............................................................ 153

V. EFFECT OF EXTRACTION PH AND TRANSGLUTAMINASE

CROSS-LINKING ON PHYSICOCHEMICAL AND TEXTURAL

CHANGES IN MYOFIBRILLAR PROTEINS EXTRACTED FROM

CHANNEL CATFISH BY-PRODUCTS....................................................... 154

5.1 Abstract ............................................................................................. 154

5.2 Introduction ....................................................................................... 155 5.3 Materials and methods ....................................................................... 156

5.3.1 Materials ...................................................................................... 156 5.3.2 Chemicals .................................................................................... 157

5.3.3 Myofibrillar protein extraction ..................................................... 157 5.3.4 Circular dichroism (CD) ............................................................... 157

5.3.5 Effect of pH on solubility of extracted myofibrillar proteins ......... 158 5.3.6 Production of protein gel and transglutaminase treatment ............. 158

5.3.7 Dynamic rheological measurement ............................................... 159 5.3.8 Differential scanning calorimetry (DSC) ...................................... 160

5.3.9 Textural analysis and cooking loss ............................................... 160 5.3.10 Color ............................................................................................ 160

5.3.11 SDS-PAGE analysis ..................................................................... 161 5.3.12 Scanning electron microscopy (SEM)........................................... 161

5.3.13 Statistical analysis ........................................................................ 162 5.4 Results and discussion ....................................................................... 162

5.4.1 Yield of extracted myofibrillar proteins ........................................ 162 5.4.2 Circular dichroism of extracted myofibrillar proteins.................... 163

5.4.3 Effect of pH on solubility of extracted myofibrillar proteins ......... 165 5.4.4 Rheological properties of protein gel made from extracted

myofibrillar proteins treated with and without

transglutaminase .......................................................................... 167

5.4.5 Differential scanning calorimetry ................................................. 171

ix

5.4.6 Penetration test, cooking loss and color of transglutaminase

treated protein gels ....................................................................... 173

5.4.7 SDS-PAGE of TGase-treated protein gel ...................................... 177 5.4.8 Scanning electron microscopy ...................................................... 178

5.5 Conclusion ......................................................................................... 181

VI. SUMMARY OF ACHIEVEMENTS, LIMITATIONS AND FUTURE

WORK .......................................................................................................... 182

6.1 Summary ........................................................................................... 182

6.2 Limitations and future work ............................................................... 183

REFERENCES............................................................................................................ 185

x

LIST OF TABLES

3.1 Kinetic constant and R2 for hydroxyproline recovery rate under

different extraction conditions. ................................................................... 59

4.1 Proximate analysis of the mixture of catfish heads and frames

(w:w=3:2). .................................................................................................. 89

4.2 Activity and cost of eight types of proteases. ............................................... 92

4.3 Hydrolysis kinetics of 8 enzymes at different temperatures and

enzyme concentrations. ............................................................................. 113

4.4 Calculated enzyme activity and cost to reach 15 and 20% DH when

hydrolyzing catfish by-product at pH 7.2. ................................................. 121

4.5 ORAC values (µmol TE/g) of selected catfish by-products

hydrolysates hydrolyzed under different conditions. ................................. 143

4.6 DPPH values (µmol TE/g) of selected catfish by-products

hydrolysates hydrolyzed under different conditions. ................................. 144

4.7 IC50 values (mg/mL) of catfish by-products hydrolysates hydrolyzed

under different conditions against angiotensin-converting enzyme. ........... 148

4.8 IC50 values (mg/mL) of selected catfish by-products hydrolysates

hydrolyzed under different conditions against DPP-IV. ............................. 149

4.9 Protein recovery rate (%) of hydrolysates with good functionalities. ......... 152

5.1 Textural analysis, cooking loss and color of TGase treated protein

gels. .......................................................................................................... 176

xi

LIST OF FIGURES

2.1 Annual catfish sales in the United States from 2010 to 2016 (USDA,

2010, 2011, 2012, 2013, 2014, 2015 & 2016). .............................................. 8

2.2 Top ten seafood consumed in the United States (National Fisheries

Institute, 2017).............................................................................................. 9

2.3 Catfish fillet processing. ............................................................................. 10

2.4 By-products from catfish fillet processing plant. ......................................... 11

2.5 SDS-PAGE patterns of collagen extracted from the skin of bigeye

snapper with acid pre-swelling process, followed by the treatment

with bigeye snapper pepsin (BSP) or porcine pepsin (PP) at 20

KUnits/g defattd skin for different times (Nalinanon, Benjakul,

Visessanguan, & Kishimura, 2007). ............................................................ 14

2.6 CD spectra of bone (a) and scale (b) collagen (Ogawa, Portier,

Moody, Bell, Schexnayder & Losso,2004). ... Error! Bookmark not defined.

2.7 CD spectra of triple helix. ........................................................................... 17

2.8 3-D structure of collagen (Goodsell, Dutta, Zardecki, Voigt, Berman

& Burley, 2000). ......................................................................................... 18

2.9 Peptide mapping of collagen from calf (lane 2) and grass carp (lane 3)

skin (Zhang, Liu, Li, Shi, Miao, & Wu, 2007)............................................. 19

2.10 UV-Vis spectra of pepsin-soluble collagen (Liu, Li, & Guo, 2007). ............ 20

2.11 Fourier transform infrared spectrum of acid-solubilized collagen of

Channel catfish skin (Liu, Li & Guo, 2007). ............................................... 21

2.12 Flow chart of surimi manufacturing (Park, 2013). ....................................... 30

2.13 Processing flow diagram of fish ball (Park, 2013). ...................................... 31

2.14 SDS-PAGE gel of sarcoplasmic proteins extracted from sea bass

(Ladrat, Verrez-Bagnis, Noël, & Fleurence, 2003) ...................................... 32

xii

2.15 Head induced gel formation of myosin monomer. ....................................... 35

2.16 Microstructure of hairtail surimi proteins (Hu, et al., 2018). ........................ 37

2.17 Microstructure of transglutaminase-treated threadfin bream surimi

gel. ............................................................................................................. 37

2.18 Microstructure of Alaska Pollock surimi treated with different

deacetylation of Konjac glucomannan (Zhang, Xue, Li, Wang, &

Xue, 2015). ................................................................................................. 38

3.1 Hydroxyproline recovery rate of collagen extracted with different

acids. .......................................................................................................... 52

3.2 Hydroxyproline recovery rate of collagen extracted with different

acids at pH 2.6. ............................................. Error! Bookmark not defined.

3.3 SDS-PAGE gels of collagens extracted with various acid extractions. ........ 62

3.4 SDS-PAGE gel of collagens extracted with different concentrations

of pepsin. .................................................................................................... 63

3.5 Effect of pH and NaCl on collagen solubility. ............................................. 66

3.6 Physical and structural characterizations of extracted collagen. ................... 70

3.7 Research diagram of utilization of catfish skin and characterization of

collagen extracted from catfish skin. ........................................................... 75

4.1 Enzymatic hydrolysis of catfish by-product (mixture of frames and

heads) with papain under different temperature and varying activity

units (AzU/g of protein in substrate). .......................................................... 96

4.2 Enzyme hydrolysis of catfish by-product (mixture of frames and

heads) with ficin under different temperature and varying activity

units (AzU/g of protein in substrate). .......................................................... 98

4.3 Enzyme hydrolysis of catfish by-product (mixture of frames and

heads) with bromelain under different temperature and varying

activity units (AzU/g of protein in substrate). ............................................ 100

4.4 Enzyme hydrolysis of catfish by-product (mixture of frames and

heads) with neutrase under different temperature and varying activity

units (AzU/g of protein in substrate). ........................................................ 102

xiii

4.5 Enzyme hydrolysis of catfish by-product (mixture of frames and

heads) with alcalase under different temperature and varying activity

units (AzU/g of protein in substrate). ........................................................ 104

4.6 Enzyme hydrolysis of catfish by-product (mixture of frames and

heads) with protamex under different temperature and varying

activity units (AzU/g of protein in substrate). ............................................ 106

4.7 Enzyme hydrolysis of catfish by-product (mixture of frames and

heads) with Novo-ProD under different temperature and varying

activity units (AzU/g of protein in substrate). ............................................ 108

4.8 Enzyme hydrolysis of catfish by-product (mixture of frames and

heads) with thermolysin under different temperature and varying

activity units (AzU/g of protein in substrate). ............................................ 110

4.9 Catfish protein and peptide pattern changes upon Protamex hydrolysis

at 50°C. .................................................................................................... 125

4.10 Catfish protein and peptide pattern changes upon Novo-proD

hydrolysis at 60°C. ................................................................................... 126

4.11 Catfish protein and peptide pattern changes upon Papain hydrolysis at

40°C. ........................................................................................................ 127

4.12 Catfish protein and peptide pattern changes upon Acalase hydrolysis

at 30°C. .................................................................................................... 128

4.13 Catfish protein and peptide pattern changes upon Neutrase hydrolysis

at 30°C. .................................................................................................... 129

4.14 Catfish protein and peptide pattern changes upon Thermolysin

hydrolysis at 50°C. ................................................................................... 130

4.15 Catfish protein and peptide pattern changes upon ficin hydrolysis at

60°C. ........................................................................................................ 131

4.16 Emulsifying activity index (EAI) and emulsion stability index (ESI)

of catfish by-products hydrolysates prepared using Novo-ProD (A

and B) and thermolysin (C and D) at 30 and 60°C, respectively. ............... 134

4.17 Foaming property and stability of catfish by-products hydrolysates. ......... 139

4.18 Research diagram of utilization of catfish by-product for making

protein hydrolysates. ................................................................................. 153

xiv

5.1 Circular dichroism spectrum of myofibrillar proteins extracted under

different pH. ............................................................................................. 165

5.2 Effect of pH on the solubility of myofibrillar proteins extracted under

different pH. ............................................................................................. 167

5.3 Storage modulus (G’) of protein sol made from myofibrillar proteins

extracted under different pH. .................................................................... 170

5.4 Storage modulus (G’) of transglutaminase treated protein sol made

from myofibrillar proteins extracted from catfish by-products

(mixture of frames and heads). .................................................................. 171

5.5 Differential scanning calorimetry thermograms of transglutaminase-

treated protein gels made from myofibrillar proteins extracted from

catfish by-products. .................................................................................. 173

5.6 SDS-PAGE gel of proteins extracted from catfish by-products, and

TGase- treated protein gel. ........................................................................ 178

5.7 Scanning electron microscopy images of TGase-treated protein gels. ........ 180

5.8 Pictures of commercial fish ball and protein gels made from catfish

by-products with (1 U/g) and without transglutaminase treatment. ............ 181

1

CHAPTER I

INTRODUCTION

Channel catfish (Ictalurus punctatus) raising is the most important aquaculture in

the United States. Channel catfish is a warm water species, native to major Mississippi

river and Gulf Coast streams in the central and southern areas of United State.

Mississippi State ranks No.1 in catfish production with annual production of $214 million

in 2017 (MSU extension data). Catfish fillet production is vital to the success of the

economy of Southern states including Alabama, Louisiana and Arkansas. The major by-

products from the fillet processing plants are heads, bone frames, and skins. Large

number of by-products (around 60% of whole catfish) generated from the catfish fillet

processing are regarded as waste and sold to local farmers for only 2 cents per pound.

Moreover, the waste from the fillet processing plants must be disposed, which otherwise,

would create environmental pollution. Therefore, to find a way to utilize by-products is

imperative.

Channel catfish skin accounts for about 10-15% of the by-products (3-6% of the

whole fish weight) (based on our preliminary study) and can be used as a potential

collagen source. Collagen is a long cylindrical protein, and is the major component that

contributes to the unique physiological function of connective tissues in the skin.

Commercial gelatin is usually extracted from porcine skins and bones. However, it is not

acceptable by Judaism and Islam due to religious restrictions (Nalinanon, Benjakul,

2

Visessanguan, & Kishimura, 2007). In addition, collagen extracted from bovine might be

contaminated with bovine spongiform encephalopathy (Choi & Regenstein, 2000).

Therefore, aquatic sources for collagen production are a substitution for mammalian

sources even though the yield of collagen from aquatic sources is much lower than that

from mammalian sources. However, the yield of collagen extracted from fish skin has

been greatly improved in recent years (Gómez-Guillén, Giménez, López-Caballero, &

Montero, 2011). More and more studies about different alternative sources and new

functionalities of collagen have been reported in the last 10 to15 years. However, most

studies used a single set of extraction conditions. There is a lack of a systematic approach

for collagen extraction, particularly for catfish collagen extraction. The extraction yield

of collagen from fish skins still remains low, and the extraction methods are of little

practical significance to the food industries since most of them require long time, high

energy input and dialysis processing. In addition, no systematic kinetic analysis of

extraction yield has been performed.

The by-products (mostly heads and bone frames) are an excellent source of

protein (42.7% protein, db) with high nutritive value, which can be processed into value-

added food products or ingredients with desirable functional and nutritional properties

(Gehring, Gigliotti, Moritz, Tou, & Jaczynski, 2011; Gehring, Davenport, & Jaczynski,

2009). Because of the steady increase in world population, global need of protein source

is increasing rapidly (Tilman, Balzer, Hill, & Befort, 2011). Catfish protein isolated from

by-products can help alleviate world protein shortage problem, and hence help reduce

problems associated with world hunger and malnutrition. Successful utilization of the by-

product as a protein hydrolysate may increase aquaculture profitability and sustainability.

3

Protein hydrolysates have been made from various fish species by bacterial

fermentation (Kristinsson & Rasco, 2000a; Sodini, Lucas, Tissier, & Corrieu, 2005) or by

enzymatic hydrolysis (Hathwar, Bijinu, Rai, & Narayan, 2011; Nilsang, Lertsiri,

Suphantharika, & Assavanig, 2005). Flavourzyme™ and Kojizyme™ have been used for

hydrolyzing tuna processing by-product (Nilsang, Lertsiri, Suphantharika, & Assavanig,

2005). Alcalase and neutrase have been utilized for hydrolyzing pacific whiting solid

waste and capelin (Benjakul & Morrissey, 1997; Shahidi, Han, & Synowiecki, 1995).

Plant proteases such as papain, ficin and bromelain have also been used for hydrolyzing

fish protein (Beddows & Ardeshir, 1979; Salampessy, Phillips, Seneweera, &

Kailasapathy, 2010). Early research intention is to use the fish hydrolysate for feed.

Recently, hydrolysates of fish protein have been found to possess antioxidant activity

(Halldorsdottir, Kristinsson, Sveinsdottir, Thorkelsson, & Hamaguchi, 2013; Mendis,

Rajapakse, & Kim, 2005), antiproliferative activity (Hsu, Chan, & Jao, 2011) and

Angiotensin-I converting enzymes (ACE) inhibition activity, which can prevent

hypertension (Nakajima, Yoshie-Stark, & Ogushi, 2009; Raghavan & Kristinsson, 2009),

and most of the studies have focused on improving the degree of hydrolysis and

antioxidant activity. However, no literature has reported the hydrolysis of Channel catfish

by-product by using proteases.

Traditionally, surimi is made from fish meat of both salt and fresh water origins.

China is the first country to produce fish ball dating back to year 220 BC (Park, 2013). In

the United States, crabsticks are made from Alaska Pollock or Pacific Whiting after

extensive washing. However, Channel catfish has never been used commercially to make

surimi or surimi-like products since it has higher value in the form of fillet. Using Alaska

4

pollock, Pacific whiting and Pacific hake meat to make surimi products such as fish ball,

fish cake and fish tofu have been studied thoroughly (Jenkelunas & Chan, 2018; Moon,

Yoon, & Park, 2017). Protein isolate from catfish fillet has been extracted by acid and

alkaline extraction methods and made into protein gel (Davenport & Kristinsson, 2011).

However, such protein extracted from catfish fillet has remained only in the stage of

research since using fillet to make surimi is cost prohibitive. In addition, the nativity of

the proteins is affected by the extraction/processing technology that imposes shearing

force during grinding and mixing, and degradation by endogenous hydrolytic enzymes,

which need to be controlled by cooling or the use of protease inhibitors. It is a challenge

to maximize processing yield and the quality of protein end-products from less desirable

raw materials such as fish by-products. Research for making food-grade products from

the by-products is still in its infancy.

Overall, catfish by-products have the potential to be processed into value added

products, for both increasing profitability and decreasing environmental pollution. An

economic analysis of producing surimi from catfish by-products shows that if extraction

rate is larger than 20% of the frame, the extraction would be commercially feasible

(McAlpin, Dillard, Kim, & Montanez, 1994). However, no comprehensive study on

utilization of catfish by-products has been reported. This study will provide a systematic

information for the catfish fillet processing industry to utilize the processing by-products.

The objectives of the present study are: (1) to use a systematic approach to

optimize the collagen extraction condition and modify the extraction method to recover

the maximum yield, (2) to characterize the properties of collagen extracted with different

extraction conditions; (3) to investigate the kinetics of the enzymatic hydrolysis of

5

proteins in catfish heads and frames, (4) to study the functional properties of enzymatic

hydrolysates, (5) to study the secondary structure changes of myofibrillar proteins

extracted under different pH, (6) to understand the physicochemical, textural and thermal

properties of the protein gel made from proteins extracted from catfish by-products, and

(7) to study the effect of transglutaminase on protein gels.

6

CHAPTER II

LITERATURE REVIEW

2.1 Aquaculture in the United State

Aquaculture is one of the most important compositions of US agriculture. Interest

in aquaculture is rising because the decreasing in the wild catch of many animal species.

US aquaculture’s estimated sales was $1.37 billion in the year of 2013 (USDA, 2013).

Catfish and trout are the major products. Most of the trout grown for food are located in

Idaho, which accounts for over 75% of the total trout production. And southern states like

Mississippi, Alabama, Louisiana and Arkansas are the major states for catfish

aquaculture. Mississippi state ranks number one in catfish production for over 20 years

(USDA, 1997-2018).

2.2 Overview of catfish production

Channel catfish farming in the Southern region is the most important warm water

aquaculture in the United State. Mississippi ranks number one in US catfish production,

which had dramatically reduced in the last 10 years due to a continuous reduction in

profitability which was caused by the increases in feed cost, and the fierce international

competition of catfish import from Vietnam. Despite the decline, catfish farming and

fillet processing industries remain as an important agricultural and food industry, and

their existence is vital to the success of the rural economy since they provide employment

opportunities to many low-income families on top of profitability to the growers. A small

7

catfish fillet processing plant can easily employ more than 250 workers. Therefore, value-

added utilization of the by-products is of a top and urgent priority to enhance the survival

of the US catfish industry.

Catfish has never been used commercially to make surimi products since it has

higher value in the form of fillet. Catfish fillet has been studied to make protein isolate by

acid and alkaline extractions, and the protein isolate was used for protein gel making

(Davenport & Kristinsson, 2011). However, such surimi or surimi-like product made

from catfish fillet has remained only in the stage of research since using fillet to make

surimi is cost prohibitive.

2.3 Overview of sales and consumption in US.

2.3.1 Catfish sales in US

Catfish productions and sales reached to $379 million in 2017 (USDA, 2018). The

combined production and sales in the southeastern states, including Alabama ($114

million), Arkansas ($20.6 million), Mississippi ($214 million), and Texas ($15.6 million),

all of which account for over 96% of the U.S. catfish production and sales in 2017

(USDA, 2018). Similar sales data were observed in the past five years. Figure 2.1

indicated the total catfish sales from 2010 to 2016.

8

Figure 2.1 Annual catfish sales in the United States from 2010 to 2016 (USDA, 2010,

2011, 2012, 2013, 2014, 2015 & 2016).

2.3.2 Catfish consumption in the US

During the past decade, shrimp is the most preferred seafood in the US. However,

catfish dropped from sixth to ninth place from 2007 to 2016. Figure 2.2 indicated the top

ten seafood consumed in the United States from 2007 to 2016. Shrimp, salmon and

canned tuna were the top three during the past decade. In addition, Alaska pollock and

tilapia are more preferred than catfish during the past ten years. Moreover, pangusius and

cod are more preferred than catfish from 2013 to 2016. These observations indicated that

the preference for seafood have been changed over the last decade in the United States,

and catfish fillet is not as preferred as before.

9

Figure 2.2 Top ten seafood consumed in the United States (National Fisheries

Institute, 2017).

2.4 Catfish fillet processing

2.4.1 Removal of skin, heads, frames and gut

In the catfish fillet processing plant, head and skin are first removed, and then the

knife will cut upward to obtain the fillet. Figure 2.3 shows the processing of catfish fillet

processing. The by-products, including heads, frames, skins, and viscera are regarded as

waste after fillet processing.

2.4.2 By-products from catfish fillet processing

Based on the US dietary preference, fish heads, frames, skins and viscera are not

acceptable to the consumers. However, depending on the type of fish, these parts can be

processed into foods and have been very popular in Asia for a long period of time. The

picture of by-products from catfish fillet processing are shown in Figure 2.4.

10

Figure 2.3 Catfish fillet processing.

11

Figure 2.4 By-products from catfish fillet processing plant.

A1 and A2: Catfish head; B: catfish skin; C1 and C2: catfish frame. (Photos are

contributed by Dr. Sam K.C. Chang).

2.5 Overview of catfish by-products

Catfish (approximately $1 per lb of the fresh fish supply) fillet processing

generates by-products, which are generally sold to rendering plants at 2 cents/lb, to

produce fish meal. The entire catfish aquaculture harvested 320.174 million pounds of

live catfish in 2016 (USDA, 2017). In catfish fillet processing, approximately 60% of the

whole catfish is the by-product, which would translate into more than 190 million pounds

of by-products. Since Mississippi growers produced 54% of all live catfish sales, catfish

processing by-products in 2016 was more than 100 million pounds.

An in-depth characterization of the physical and biochemical properties of the

catfish by-products as affected by processing and engineering methods has not been

researched. Such characterization is needed for scale-up production. The fish by-products

can undergo rapid changes (autolysis by endogenous proteases) to negatively affect the

12

quality and yield of the extracted proteins. Component fractionation (based on water

solubility, pH and salt concentrations) and understanding the molecular and biological

characteristics of the fractions are of the primary importance for developing novel

integrated technologies for the complete utilization of the catfish by-products.

2.6 Fish skin

Channel catfish skin accounts for about 3-6% of the whole fish weight (10-15%

of the by-products). Collagen is the major components of skin and contributes to the

unique physiological function of connective tissue in the skin. The skin contains mostly

connective proteins (collagen) and fat. Collagen had been isolated from fish skin and

scale by researchers (Chen, Li, Yi, Xu, Gao, & Hong, 2016; Chen, et al., 2016; Duan,

Zhang, Du, Yao, & Konno, 2009; Huang, Kuo, Wu, & Tsai, 2016). Collagen is

traditionally extracted from the mammalian sources such as porcine and bovine; however,

it is not acceptable by Judaism and Islam due to religious restrictions (Nalinanon,

Benjakul, Visessanguan, & Kishimura, 2007). Therefore, aquatic sources for collagen

production are a substitution for the mammalian sources even though the yield of

collagen from aquatic sources is much lower than that from mammalian sources.

Fortunately, the yield of collagen extracted from fish skin has been greatly improved in

the recent years (Gómez-Guillén, Giménez, López-Caballero, & Montero, 2011).

Researchers have reported that around 50% of collagen can be isolated from fish skin,

and 45% of collagen from fish bones (Nagai & Suzuki, 2000).

13

2.6.1 Extraction and characterization of collagen

2.6.1.1 Extraction method

There are three major methods for collagen extraction: acid extraction, enzyme-

aided extraction and alkaline extraction, among which enzyme-aided extraction and acid

extraction methods are the most commonly used (Nalinanon, Benjakul, Visessanguan, &

Kishimura, 2007; Tang, Chen, Su, Weng, Osako, & Tanaka, 2015). Researchers found

that the molecular weights of collagen extracted from fish skin with different methods

(enzyme-aided method and acid extraction method) exhibited a slight difference. The

molecular weight of collagen extracted with enzyme-aided (pepsin) method was a little

bit lower than that from the acid extraction method (Figure 2.5). In addition, the

structural changes induced by pepsin extraction may have some effects on the thermal

stability and solubility of collagen (Nalinanon, Benjakul, Visessanguan, & Kishimura,

2007). Denaturation temperatures of collagen from fish skin, bones, and fins were found

to be 9°C lower than that from porcine (Nagai & Suzuki, 2000). Collagen had been

extracted from three species, and further filtered through a phosphocellulose column into

two major fractions (type I and type V collagen). Peptide mapping, and amino-acid

analysis of these two fractions indicated that type I collagen was widely present in the

fish skin as a major collagen (Yata, Yoshida, Fujisawa, Mizuta, & Yoshinaka, 2001).

14

Figure 2.5 SDS-PAGE patterns of collagen extracted from the skin of bigeye snapper

with acid pre-swelling process, followed by the treatment with bigeye

snapper pepsin (BSP) or porcine pepsin (PP) at 20 KUnits/g defattd skin

for different times (Nalinanon, Benjakul, Visessanguan, & Kishimura,

2007).

2.6.1.2 Secondary structure of collagen

Up to 27 types of collagen have been identified, and type I collagen exists the

most widely in the connective tissues. Total molecular mass of collagen is about 300

KDa with each chain has a molecular mass of about 100 KDa, and it has a wide range of

applications in pharmaceutical, leather, biomedical and film industries (Ogawa, Portier,

Moody, Bell, Schexnayder, & Losso, 2004).

Circular dichroism (CD) spectra of collagen extracted from black drum and

sheepshead seabream with acid and pepsin-aided methods were reported by Ogawa and

coworkers. They found native collagen from bones gave a characteristic CD spectrum

with a positive extreme at 220 nm and a negative peak appeared at 197–199 nm (Figure

2.6a). Slight deviations in ellipticity were observed among the four measured collagen,

suggesting that there was a minor discrepancy in molecular structures. On the other hand,

15

scale collagens showed similar secondary structures, as seen in Figure 2.6b, denoting a

positive extreme at 220 nm and a negative peak at 198 nm (Ogawa, Portier, Moody, Bell,

Schexnayder, & Losso, 2004). Greenfield (2006) has reported the triple helix structure of

collagen and denatured collagen using circular dichroism (Figure 2.7). α-Helical proteins

have negative peaks at 222 nm and 208 nm and a positive band at 193 nm (line 4),

whereas disordered proteins (line 5) have very low ellipticity above 210 nm and negative

peaks near 195 nm.

Figure 2.6 CD spectra of bone (a) and scale (b) collagen (Ogawa, Portier, Moody,

Bell, Schexnayder & Losso,2004).

CD spectra were taken at 15°C. Filled diamond: black drum acid soluble collagen;

unfilled diamond: black drum pepsin soluble collagen; filled triangle, sheepshead acid

soluble collagen; unfilled triangle: sheepshead pepsin soluble collagen (Ogawa, Portier,

Moody, Bell, Schexnayder, & Losso, 2004).

16

Figure 2.6 (continued)

CD spectra were taken at 15°C. Filled diamond: black drum acid soluble collagen;

unfilled diamond: black drum pepsin soluble collagen; filled triangle, sheepshead acid

soluble collagen; unfilled triangle: sheepshead pepsin soluble collagen (Ogawa, Portier,

Moody, Bell, Schexnayder, & Losso, 2004).

17

Figure 2.7 CD spectra of triple helix.

(4) triple helix structure of collagen and (5) denatured collagens (Greenfield, 2006).

2.6.1.3 3-D structure of collagen

Collagen is a long cylindrical protein, and Type I collagen is made up of three

polypeptide chains, two of the polypeptides are designated α1, and α1 bonded to another

chain to form a third chain α2 through hydrogen bonds. The 3-D structure of collagen is

shown in Figure 2.8. Peptides in the telopeptide region could be cleaved by pepsin with

limited pepsin concentration, and the cross-linked molecules at the telopeptide region can

be cleaved by pepsin but without damage the triple helix structure of collagen (Liu, Li,

Miao & Wu, 2009).

18

Figure 2.8 3-D structure of collagen (Goodsell, Dutta, Zardecki, Voigt, Berman &

Burley, 2000).

2.6.1.4 Peptide mapping and amino acid composition of collagen

Peptide mapping and amino acid composition of collagen are very common

characteristics that researchers often focus on. Peptide mapping usually gives very similar

results for collagen extracted from different parts of the same specie of fish (Nagai,

Araki, & Suzuki, 2002; Zhang, Liu, Li, Shi, Miao, & Wu, 2007). However, collagen from

different species of fish exhibied different peptide maps. Figure 2.9 shows the one-

dimmentional SDS-PAGE peptide mapping of collagen extracted from calf and grass

carp skins, which suggested that collagens extracted from calf and grass carp skins were

different. The band intensity of β/(α1+ α2) chain from collagen extracted from porcine or

bovine has not been reported before. The amino acid composition of collagen was firstly

19

reported back to the year of 1955 (Eastoe, 1955). The major amino acid is glycine and the

amino acid composition exhibits slight differences due to different speices (Muyonga,

Cole, & Duodu, 2004a; Piez & Gross, 1960). In addition, hydroxyproline is a major

component of collagen and plays a key role in the stability of the triple helix structure of

collagen. It can be used as an indicator to determine the amount of collagen.

Figure 2.9 Peptide mapping of collagen from calf (lane 2) and grass carp (lane 3) skin

(Zhang, Liu, Li, Shi, Miao, & Wu, 2007).

2.6.1.5 Fourier transform infrared spectrum of collagen

UV-Vis absorption sepectra and fourier transform infared spectroscopy were also

applied to fish skin collagen to study the absorption characteritics and spectra. The UV-

Vis spectra of pepsin-solublized collagen is shown in Figure 2.10, the biggest absorption

of the skin collagen of catfish was obtained near 232 nm (Liu, Li, & Guo, 2007).

20

Spectrometer simultaneously collects high spectral resolution data over a wide spectral

range. Figure 2.11 shows the fourier transform infrared spectroscopy of the acid-

solublized collagen from the catfish skin, and it is similar to other species collagens

(Muyonga, Cole, & Duodu, 2004b).

Figure 2.10 UV-Vis spectra of pepsin-soluble collagen (Liu, Li, & Guo, 2007).

21

Figure 2.11 Fourier transform infrared spectrum of acid-solubilized collagen of

Channel catfish skin (Liu, Li & Guo, 2007).

2.6.1.6 Solubility of extracted collagen

Collagen solubility is an important characteristic that researchers often focus on.

It has been reported that the pI of collagen varies from 6 to 9 (Foegeding, Lanier, &

Hultin, 1996). Collagen extracted with pepsin from bigeye snapper has been reported to

have the highest solubility at pH 5 (Nalinanon, Benjakul, Visessanguan, & Kishimura,

2007). The solubilities of collagen extracted from the skins of bigeye snapper, striped

catfish, and brownstripe red snapper (Jongjareonrak, Benjakul, Visessanguan, Nagai, &

Tanaka, 2005; Kittiphattanabawon, Benjakul, Visessanguan, Nagai, & Tanaka, 2005; P

Montero, Gómez-Guillén, & Borderıas, 1999; Singh, Benjakul, Maqsood, & Kishimura,

2011) decrease with the increasing concentration of sodium chloride. When a higher

concentration of sodium chloride (> 3%) is present, an increasing ionic strength might

lead to a reduction of protein solubility by increasing the interactions between protein

22

chains. Thus, the solubility of protein might be decreased by salting out effect via

increasing hydrophobic interaction and aggregation.

2.6.1.7 Gel strength of collagen-induced gel

Gel strength is one of the most important indexes for collagen quality and it could

be classified into 3 levels: low (<150 g), medium (150-220 g) and high (220-300 g) gel

strength (Johnston-Bank, 1983). The strength of the collagen-converted gel with the same

concentration have been reported for red tilapia skin (128.11 g) (Jamilah & Harvinder,

2002), Atlantic salmon (108 g), cod (71 g) (Arnesen & Gildberg, 2007), young and adult

Nile perch skins (222 and 229 g, respectively), young and adult Nile perch bone (179 and

134 g, respectively) (Muyonga, Cole, & Duodu, 2004b), shortfin scad (177 g) (Cheow,

Norizah, Kyaw, & Howell, 2007), and tilapia scale (157- 260 g) (Huang, Kuo, Wu, &

Tsai, 2016). Commercial pork skin gelatin showed higher gel strength than that of catfish

skin (Avena‐Bustillos et al., 2006). The discrepancy in gel strength among species might

be due to the different amino acid composition and molecular size of protein chains

(Muyonga, Cole, & Duodu, 2004b). When gel strength was measured at low temperature

(< 10 °C), some short chain peptides present in low viscosity gelatins tend to strengthen

the gel (Montero & Gómez-Guillén, 2000). In addition, the thermal shrinkage,

denaturation temperature of collagen and melting temperature of gelatins isolated from

cold-water fish are significantly lower than those from fish living in warm waters. The

reason of the differences might be due to a lower degree of proline hydroxylation of cold-

water fish collagen (Gilsenan & Ross-Murphy, 2000).

23

2.7 Heads and frames

2.7.1 Overview

The by-products (mostly heads and bone frames) are an excellent source of

protein (42.7% protein, db) with high nutritive value, and the protein can be processed

into value-added food products or ingredients with desirable functional and nutritional

properties (Gehring, Gigliotti, Moritz, Tou, & Jaczynski, 2011; Gehring, Davenport, &

Jaczynski, 2009). However, heads and frames are not collected separately in some fillet

processing plants, and therefore, to develop a collection logistics of heads and frames is

essential to help industries utilize the rich proteins in these by-products. At the meantime,

heads and frames are the rich sources of calcium. Calcium based powder had been

produced from Alaska pollack bones (Choi, Kim & Kim, 1998).

2.7.2 Hydrolysis of by-products

Many researchers have focused research on the hydrolysis of aquaculture by-

products for reducing environmental pollution consideration. Protein hydrolysates have

been made from various fish species by bacterial fermentation (Kristinsson & Rasco,

2000a; Sodini, Lucas, Tissier, & Corrieu, 2005) or by enzymatic hydrolysis (Hathwar,

Bijinu, Rai, & Narayan, 2011; Nilsang, Lertsiri, Suphantharika, & Assavanig, 2005).

Flavourzyme™ and Kojizyme™ have been used for hydrolyzing tuna processing liquid

waste (fish soluble concentrate) (Nilsang, Lertsiri, Suphantharika, & Assavanig, 2005).

Alcalase and neutrase have been utilized for hydrolyzing pacific whiting solid waste and

capelin (Benjakul & Morrissey, 1997; Shahidi, Han, & Synowiecki, 1995). Plant

proteases such as papain, ficin and bromelain have also been used for hydrolyzing

leatherjacket proteins (Salampessy, Phillips, Seneweera, & Kailasapathy, 2010). Early

24

research goal was to use the fish hydrolysates for pet feed. Recently, hydrolysates of

herring proteins have been found to possess antioxidant activity (Halldorsdottir,

Kristinsson, Sveinsdottir, Thorkelsson, & Hamaguchi, 2013; Mendis, Rajapakse, & Kim,

2005), antiproliferative activity (Hsu, Chan, & Jao, 2011) and angiotensin-I converting

enzymes (ACE) inhibitory activity, which can prevent hypertension (Nakajima, Yoshie-

Stark, & Ogushi, 2009; Raghavan & Kristinsson, 2009). However, most of the studies

have focused on improving the degree of hydrolysis and antioxidant activity.

The most important control points of hydrolyzing fish by-products are the degree

of hydrolysis and bitterness, which usually caused by some specific amino acids such as

tryptophan and hydrophobic amino acids. Therefore, to help reach a high degree of

hydrolysis and low level of bitterness, different enzymes have been used to hydrolyze

fish by-products such as Flavourzyme and Kojizyme. It was found that Kojizyme

enhanced the formation of bitterness during hydrolysis whereas Flavourzyme did not

(Nilsang, Lertsiri, Suphantharika, & Assavanig, 2005). The protein content of

hydrolysate fermented with Flavourzyme was found to be higher than that with Kojizyme

(Nilsang, Lertsiri, Suphantharika, & Assavanig, 2005). Flavourzyme and savorase have

been used to hydrolyze red hake by-products, and the results showed that the addition of

sodium chloride and sodium tripolyphosphate (STPP) improved the flavor quality and

limited the bitterness and off-flavor (Imm & Lee, 1999). Alcalase was used for

fermentation of tuna by-products, and a linear relationship was found between enzyme

concentration and the degree of hydrolysis (Guerard, Dufosse, Broise, & Binet, 2001).

In areas of whole fish waste utilization, protein hydrolysates have been made by

bacterial fermentation (Murthy, Rai, & Bhaskar, 2014) or by enzymatic hydrolysis

25

(Mackie, 1974). Papain and alcalase have been used to hydrolyze fish proteins (Amiza,

Nurul Ashikin, & Faazaz, 2011; Hoyle & Merritt, 1994). Other examples of protein

hydrolysates include hydrolysates from sardine waste by pepsin (Benhabiles, et al.,

2012), from croaker waste by fungal protease and alcalase (Hathwar, Bijinu, Rai, &

Narayan, 2011) and from fish bones by Bacillus protease (Kumar & Bhalla, 2003).

Pepsin hydrolyzed fish peptides can be reformed into plastein gel (Onoue & Riddle,

1973).

2.7.3 Proteolytic activity determination

The proteolytic activity of each protease varies significantly, which is not

surprising because that proteases (derived from different sources) are expected to have

different proteolytic activities. Most of the proteases were obtained with different

proteolytic activity. However, the proteolytic activity of each protease is determined at

the optimal reaction conditions. In the literature, the substrate used for determining the

proteolytic activity is not the same, which makes it difficult to compare the effect of the

protease across various reported studies.

Some studies report the activities of the added proteases on the basis of

proteolytic activity are based on the activity provided by manufacturer (Benjakul &

Morrissey, 1997), and did not adjust the protease to the same proteolytic activity level. In

addition, the proteolytic activity has been determined using different conditions and

different substrates. Different hydrolysis conditions usually lead to different proteolytic

activity, and substrate specificity influences the protease efficiency as well.

26

2.7.4 Emulsion properties of by-product hydrolysates

Protein hydrolysates with a low degree of hydrolysis could be used as an

emulsifier in some cases. Hydrolysis influences the molecular size, hydrophobicity and

polarity of the hydrolysates (Kristinsson & Rasco, 2000a). And the characteristics of

hydrolysates would affect the functional properties and have the applications as food

ingredients directly (Kristinsson & Rasco, 2000a). Protein hydrolysates usually exhibited

better solubility over a wide range of pH, and this brings a useful characteristic for food

applications. In addition, hydrolysis also affects emulsifying and foaming properties of

the hydrolysates. A high degree of hydrolysis produces a negative effect on emulsifying

and foaming properties (Kristinsson & Rasco, 2000a). Good emulsion and foaming

properties of fish protein hydrolysates derived from sardine, salmon, and yellow strip

were observed when the degree of hydrolysis is low (Quaglia & Orban, 1990; Gbogouri,

Linder, Fanni, & Parmentier, 2004; Klompong, Benjakul, Kantachote, & Shahidi, 2007).

2.7.5 Antioxidant activity of fish protein hydrolysates

Hydrolysates of proteins are peptides with different molecular mass, and

hydrolysates of sardine and horse mackerel muscle proteins have been found to possess

antioxidant activity (Morales-Medina, Tamm, Guadix, Guadix, & Drusch, 2016).

Enzymatic hydrolysates of purple sea urchin gonad have been reported to have the ability

of scavenging DPPH radical (Qin et al., 2011). Some studies indicated that the

antioxidant activity of protein hydrolysates was positively correlated with the degree of

hydrolysis of the hydrolysates (Dong, Zeng, Wang, Liu, Zhao & Yang, 2008; Klompong,

Benjakul, Kantachote & Shahidi, 2007). Hydrolysates with higher degree of hydrolysis

have been expected to expose more hydrophobic amino acid residue side chain groups

27

and easier accessed by radical (Qin et al., 2011). The exposed hydrophobic amino acid

side chain groups would accelerate electron transfer from peptides to radical, and then

make the radical more stabilized.

2.7.6 Angiotensin-converting enzyme (ACE) inhibition activity of fish protein

hydrolysates

Angiotensin-converting enzyme (ACE), is a central component of the renin-

angiotensin system (RAS), which controls blood pressure by regulating the volume of

fluids in the body. It converts the hormone angiotensin I to the active vasoconstrictor

angiotensin II. Therefore, ACE indirectly increases blood pressure by causing blood

vessels to constrict. Hydrolysates of leatherjacket muscle protein exhibited angiotensin-I

converting enzymes (ACE) inhibition activity (Salampessy, Reddy, Phillips, &

Kailasapathy, 2017). Sardine by-product hydrolysates have been reported to have ACE

inhibition capacity with IC50 values up to 7.4 mg/mL (Bougatef et al., 2008). However,

no positive control was reported in the above literature. Regarding to the relationship

between peptides structure and ACE inhibitory activity, peptides had Pro, Phe, or Tyr at

the C-terminus, and Val and Ile at the N-terminus exhibited potent inhibitory activity

against ACE (Tsai, Chen & Pan, 2008).

2.7.7 Dipeptidyl peptidase IV (DPP- IV) inhibition activity of fish protein

hydrolysates

Dipeptidyl peptides IV (DPP-IV) plays a critical role in maintaining glucose

homeostasis. It is responsible for inactivating incretins, such as glucose-dependent

insulinotropic polypeptide (GIP) and glucagon-like peptide-1 (GLP-1). Secretion of these

two intestinal hormones is triggered by food intake. After DPP-IV was blocked by

28

inhibitor, the increased incretin level will inhibit glucagon release, which in turn

increases insulin secretion, decreases gastric emptying, and decreases blood glucose level

(Deacon, 2018). Two peptides Gly-Pro-Ala-Glu and Gly-Pro-Gly-Ala have been purified

from Atlantic salmon skin gelatin hydrolysates and exhibited potent DPP-IV inhibition

activity (IC50 = 49.6 and 41.9 μM, respectively) (Li-Chan, Hunag, Jao, Ho & Hsu, 2012).

Barbel muscle hydrolysates have been separated by HPLC into 5 fractions and two of

them showed DPP-IV inhibition activity with IC50 = 1.09 and 0.21 mg/mL, respectively.

Numerous peptides showed DPP-IV inhibitory activity have been reported in literature

(Le Maux, Nongonierma, Murray, Kelly, & FitzGerald, 2015; Nongonierma, Paolella,

Mudgil, Maqsood & FitzGerald, 2018). However, the inhibitory mechanism is remained

unclear.

2.8 Overview of viscera

Channel catfish by-product viscera has been used to extract fish oil and proteases

(Jiang, Lu, Liao, Lin, Wen & Jiang, 2009; Bougatef et al., 2008). Proteinases were

successfully isolated from Atlantic cod, cod stomach has been reported to be contained

about 2 g pepsin/kg of stomach, and the intestine contained about 1 g trypsin like

enzymes (such as trypsin, chymotrypsin and elastase) (Gildberg, 1992). Carp viscera had

been fermented and fish oil was extracted from the fermentation products, and the quality

of the extracted oil has been studied. The results indicated that the acid value of the

extracted oil was increased with the extend of fermentation time (Rai, Swapna, Bhaskar,

Halami & Sachindra, 2010).

29

2.9 Surimi

2.9.1 Historical review of surimi

Traditionally, surimi is made from fish meat of both salt and fresh water origins.

China is the first country to produce fish ball, dated back to year 220 BC (Park, 2013).

Surimi is refined fish myofibrillar proteins produced via step-by-step washing process.

Surimi processing equipment was first introduced at the end of the World War II

in 1945, and the production volume started to rise after that. With Magnuson Act on

1976, United States began the surimi production with Alaska pollock as the raw material.

There are 17-land based surimi processing plants that have been built since 1984 in US

(Park, Nozaki, Suzuki, & Beliveau, 2013).

2.9.2 Manufacture of surimi

Mechanical fish meat separator is used for deboning. The flesh is minced and then

washed and dewatered. Before the final dewatering step, impurities and pigments such as

skin, small scales and pin bones were removed by a refiner. Screw press was then to

remove extra water of minced meat, the moisture of the minced meat was adjusted to

80%-84%. Addition of cryoprotectant would protects the surimi from freezing induced

protein denaturation and aggregation. Figure 2.12 shows the flow chart of surimi

manufacturing.

30

Figure 2.12 Flow chart of surimi manufacturing (Park, 2013).

2.9.3 Manufacture of fish ball

Fish ball is a popular food in the southeast of Asia, and have different names in

different countries: Yuwan in China, Bebola Ikan in Indonesia and Malaysia, Look Chin

Pla in Thailand (Park, 2013). The raw material for making fish ball is minced fish meat.

Salt is added during chopping and usually 3-5% by weight (Kok, Thawornchinsombut &

Park, 2013). Once chopping is finished, the paste is transferred to a forming machine to

produce fish ball. Figure 2.13 shows the flow chart of fish ball manufacturing.

31

Figure 2.13 Processing flow diagram of fish ball (Park, 2013).

2.9.4 Protein component of surimi (washed fish meat mince)

The proteins in minced meat including myofibrillar proteins and less sarcoplasmic

proteins. The predominant protein is myosin (55-60% of total myofibrillar proteins) and

actin comprises 15-30% of the total myofibrillar proteins. Myosin contributes the most to

the gelation and water binding. However, myosin binds with actin tightly postmortem and

resulting in a complex protein actomyosin. Therefore, actomyosin is the predominant

32

protein in the surimi products. Other proteins related with myosin or actin including

tropomyosin, troponin complexes actinins, titin, nebulin, M-proteins and C-proteins

(Asghar & Pearson, 1980).

In addition, sarcoplasmic proteins contain many proteases, which involved in

muscle metabolism. SDS-PAGE of fish sarcoplasmic proteins showed that the protein

bands of sarcoplasmic proteins range from 12 to 97 KDa and as shown in Figure 2.14

(Ladrat, Verrez-Bagnis, Noël, & Fleurence, 2003). And it is less pH dependent than that

of myofibrillar proteins which exhibit a U-shaped solubility curve with the lowest

solubility observed at pH 5.5 (Tadpitchayangkoon, Park, & Yongsawatdigul, 2010).

Figure 2.14 SDS-PAGE gel of sarcoplasmic proteins extracted from sea bass (Ladrat,

Verrez-Bagnis, Noël, & Fleurence, 2003)

Left lane: marker; Right lane: sarcoplasmic proteins.

33

2.9.5 Myofibrillar proteins extraction method

2.9.5.1 Alkaline extraction

Fish protein extraction/solubilization by varying pH’s has been reviewed

(Kristinsson, Lanier, Halldorsdottir, Geirsdottir, & Park, 2013; Lansdowne, Beamer,

Jaczynski, & Matak, 2009; Nolsøe & Undeland, 2009). Different fish species and their

by-products respond differently to salt or alkaline extraction process due to differences in

raw material’s physical, chemistry and biological properties. Furthermore, foods are

biomaterials that exist in a biochemically dynamic state. The chemical composition,

quality of the protein and their functionalities continue to the changes during processing

and post-processing handling and storage. Proteins from catfish fillet under alkaline and

acidic conditions and followed by pH 5.5 precipitation (isoelectric precipitation) have

been reported (Davenport & Kristinsson, 2011). The gel of the protein isolated from

catfish fillet under alkaline conditions is firmer than that by acidic extraction method

(Halldórsdóttir, Hamaguchi, Sveinsdóttir, Kristinsson, Bergsson, & Thorkelsson, 2011).

Alkaline extraction is believed to unfold the myosin head structure to become more

hydrophobic which result in a firmer product after cooking. The yield of alkaline-

extracted protein is higher than the traditional surimi making method since alkaline

extraction also extracts sarcoplasmic proteins, which are water-soluble and washed off by

the traditional mincing and the extensive water-washing process (Kristinsson, Lanier,

Halldorsdottir, Geirsdottir, & Park, 2013).

2.9.5.2 Salt extraction

It is well known that functional proteins could be extracted from animal sources

or plant seeds by salt solutions. Extraction rate depends on several factors, including

34

solvent/solid mass ratio, salt concentration and stirring/blending method. Water-soluble

sarcoplasmic proteins can be extracted with repeated washing of homogenized fish with

water or dilute salt solution of less than 0.15 M NaCl (Kristinsson, Lanier, Halldorsdottir,

Geirsdottir, & Park, 2013), at which myofibrillar proteins are remain insoluble.

2.9.5.3 Acid extraction

Acid extraction (pH 2.5) method have been tested to extract proteins from

Channel catfish muscle, and no hydrolytic break down was observed during low pH

extraction. Results showed that acid extraction method produced more denaturation of

proteins, and the protein gel made from acid-extracted proteins shows significantly higher

whiteness than that of alkaline extraction. Moreover, alkaline-extraction recovers more

heme proteins than that of acid extraction (Kristinsson, Theodore, Demir, & Ingadottir,

2005).

2.9.5.4 Extraction temperature

Protein functionalities could be better retained if more native molecular structure

or extracted proteins remained after extraction process. Study has shown that bovine

muscle stored at 37°C promoted a quicker protein degradation in myofibrillar proteins

than that stored at 4°C (Yates, Dutson, Caldwell, & Carpenter, 1983). Therefore, protein

extraction should be conducted under low temperature to prevent degradation of

myofibrillar proteins (Yates, Dutson, Caldwell, & Carpenter, 1983). Moreover, fish

digestive proteases such as pepsin and pancreatin would break down the proteins at 37°C,

and low temperature would decrease the activities of these proteases.

35

2.9.6 Gelation chemistry

Previous study indicated that myosin contributes the most to gel formation and

water holding in cooked meat products. Single myosin molecule should be the first

released from muscle during chopping with salt. Salt could boost the solubilization of

myosin. Figure 2.15 shows the gel formation of myosin monomer. Aggregation of

myosin molecules during heating is initiated by disulfide cross-linking of denatured head

region of myosin, and continuous heating caused the unfolding of the tail portion of the

myosin (Yasui & Samejima, 1990).

Figure 2.15 Head induced gel formation of myosin monomer.

(a) Unheated myosin monomer; (b) formation of cluster in the early heating stage; (c)

Daisy wheel formation of myosin; (d) denaturation of tail portions; (e) clusters bound

through denatured tails; (f) gel network formation. (Park, 2013; Iwasaki, Noshiroya,

Saitoh, Okano, & Yamamoto, 2006).

36

2.9.7 Microstructure of surimi and surimi like products

Microstructure of fish surimi gel has been studied 15 years. Filamentous structure

of Walleye Pollock surimi gel was observed (Kubota, Tamura, Morioka, & Itoh, 2003).

Scanning electron microscopy with high magnification has been used to study the

microstructure of hairtail surimi proteins and the graph is shown in Figure 2.16 (Hu, et

al., 2018), which shows a discontinuous network with large grooves and holes.

Transglutaminase has been used to improve the threadfin bream surimi gel quality, and

the microstructure of transglutaminase-treated threadfin bream surimi gel is shown in

Figure 2.17 (Kaewudom, Benjakul, & Kijroongrojana, 2013), which shows myofibrillar

proteins could form a more compact and denser gel network with transglutaminase

treatment. Different deacetylations of Konjac glucomannan have been used to improve

the network of Alaska Pollock surimi gel, and microstructure of treated gel is shown in

Figure 2.18. The gel with Konjac glucomannan-treated (Day 0) shows a denser and more

uniform network structure than that of the control, and the gel shows more compact and

with smaller holes in the network when the deacetylation of Konjac glucomannan

increased (Day 2 and Day 3) (Zhang, Xue, Li, Wang, & Xue, 2015).

37

Figure 2.16 Microstructure of hairtail surimi proteins (Hu, et al., 2018).

Figure 2.17 Microstructure of transglutaminase-treated threadfin bream surimi gel.

(a) untreated surimi gel; (b) transglutaminase treated (1.2 U/g) surimi gel (Kaewudom,

Benjakul, & Kijroongrojana, 2013).

38

Figure 2.18 Microstructure of Alaska Pollock surimi treated with different

deacetylation of Konjac glucomannan (Zhang, Xue, Li, Wang, & Xue,

2015).

Da0: surimi treated with Konjac glucomannan for 0 day. Da2: surimi treated with Konjac

glucomannan for 2 days. Da3: surimi treated with Konjac glucomannan for 3 days.

39

CHAPTER III

ISOLATION AND CHARACTERIZATION OF COLLAGEN EXTRACTED FROM

CHANNEL CATFISH (ICTALURUS PUNCTATUS) SKIN

3.1 Abstract

Channel catfish skin is a by-product from catfish fillet production. Collagen was

extracted from catfish skins by: (1) acid extraction; (2) homogenization-aided; and (3)

pepsin-aided extraction methods. Kinetic analysis of the extraction was performed. SDS-

PAGE was carried out for all collagens extracted under different conditions. Protein

solubility, zeta potential, circular dichroism and gel strength of collagen extracted by

three methods were studied to determine optimal conditions. Hydroxyproline recovery

rate from minced skins extracted with pH 2.4 HCl containing 23.6 KU/g pepsin was the

highest (64.19%, db). SDS-PAGE showed that collagens extracted with different

methods had different proteins ratio patterns, even though the molecular mass of collagen

subunits was similar, 123 and 113 KDa for α1 and α2 chains, 226 KDa for β chain and

338.5 KDa for γ chain, respectively. Acid solubilized collagen, homogenization aided

solubilized collagen and pepsin homogenization solubilized collagen were made into gel,

the gel strength were 223.13, 220.48 and 73.59 ɡ, respectively. Channel catfish skin

collagens were typical type-I collagen and could have applications in food, medical and

cosmetic industries.

Key words: Channel catfish skin, collagen, homogenization, pepsin, kinetic analysis

40

3.2 Introduction

Collagen is a long cylindrical protein, it is the major protein in the skin and

contributes to the unique physiological function of the connective tissue. Up to 27 types

of collagens have been identified, and Type-I collagen exists the most widely in the

connective tissues. Type-I collagen is made up of three polypeptide chains, two of the

polypeptides are designated α1 chain, and α1 chain bonded to another chain to form a

third chain α2 through hydrogen bond. Total molecular mass of collagen is about 300

KDa with each chain has a molecular mass of about 100 KDa, and it has a wide range of

applications in the pharmaceutical, leather, biomedical and film industries (Ogawa,

Portier, Moody, Bell, Schexnayder, & Losso, 2004). Collagen usually extracted from

porcine skins and bones, however, it is not acceptable by Judaism and Islam due to

religious restrictions (Nalinanon, Benjakul, Visessanguan, & Kishimura, 2007). In

addition, collagen extracted from bovine might be contaminated with bovine spongiform

encephalopathy (BSE) (Choi & Regenstein, 2000). Therefore, collagen isolated from

aquatic sources is a substitution for mammalian sources even though the yield of the

collagen from aquatic sources is around 2-4 times lower than that from mammalian

sources. However, the yield of collagen extracted from fish skin was greatly improved in

recent years (Gómez-Guillén, Giménez, López-Caballero, & Montero, 2011).

Researchers have reported that around 50% of collagen in fish skin, and 45% in fish bone

(Nagai & Suzuki, 2000) can be isolated.

Channel catfish aquaculture is an important aquaculture in the United States, and

Mississippi State ranks No. 1 in catfish production with annual production of $214

million in 2017 (USDA, 2018). Catfish fillet processing contributes to the vitality of the

41

local economy. The fish skins account for about 10-15% of the by-products (3-6% of

whole fish weight) and can be used as a potential collagen source. More and more

studies about different alternative sources and new functionalities of collagen had been

reported in the last 10 to15 years. However, most studies used a single set of solvent

conditions for extraction. There is a lack of a systematic approach for collagen extraction,

particularly for catfish collagen extractions. And the reported yield of the collagen from

fish skins remained low, and therefore, the extraction methods have less practical

applications for the food industries since most of them require long time, high energy

input and dialysis processing. In addition, no systematic kinetic analysis of extraction

yield has been performed. Our objectives were to use a systematic approach to optimize

the extraction conditions and modify the extraction method to recover the maximum

yield, and to characterize the physical-chemical properties and gel strength of the

collagen extracted by different extraction conditions.

3.3 Materials and methods

3.3.1 Materials

Catfish skin was collected from a local catfish fillet processing (Country Select,

Isola, MS). The catfish skin was buried in crushed ice during transportation to our

laboratory and stored at -80°C until use.

3.3.2 Chemicals

All chemicals, reagents and porcine gastric pepsin (EC 3.4.23.1) were of the

analytical grade and obtained from the Sigma-Aldrich Chemical Company (St. Louis,

MO, U.S.A).

42

3.3.3 Proximate analysis

Moisture, protein, ash and total fat content were determined by the AOAC

International methods 934.01, 955.04, 942.05 and 2003.06, respectively (AOAC

International, 2012). A conversion factor of 5.4 was used for calculating the crude protein

content from nitrogen content. The hydroxyproline content in skin material and extracted

collagen was determined according to the method of Bergman and Loxley (Bergman &

Loxley, 1963). The hydroxyproline recovery rate (%) was obtained by the following

equation:

Hydroxyproline recovery rate(%) = hydroxyproline content in extracted collagen sample

hydroxyproline content in skin material×

100% (3.1)

3.3.4 Extraction of collagen with acids (ASC)

Catfish skins were pretreated by washing with iced water containing 1% NaCl

(1:6 W/V) and ground into small pieces, then passed through a 35-mesh sieve. Different

size of skins was obtained by passed the ground skin through series size of sieve (5 mesh,

10 mesh, 18 mesh and 35 mesh). One hundred grams of catfish skins were mixed with 4

different acids (acetic acid, hydrochloric acid, citric acid and lactic acid, W:V = 1:50)

with varying pH’s (1.8, 2.1, 2.4, 2.7 and 3.0), and shaken for 48 hr (100 rpm) using an

orbital shaker at 4°C, then centrifuged at 15000 ɡ for 20 min at 4°C to collect the

supernatant, the residue was re-extracted with the corresponding acid (W/V = 1:50) for

another 12 hr. After the desire time was achieved, the mixture was centrifuged at the

same conditions, the supernatant was combined and salted out by adding NaCl to a final

concentration of 0.9 M (Ogawa, Portier, Moody, Bell, Schexnayder, & Losso, 2004), the

43

precipitated collagen was separated by centrifugation at 15,000 ɡ for 15 min at 4°C. The

resultant precipitate was washed quickly in Type-I water (ultrapure water) (W/W = 1:2)

for 3 times to remove NaCl and then lyophilized.

3.3.5 Extraction of collagen with homogenization-aided (HSC) method

Preliminary experiments showed that homogenization (Model: GLH 580, Spindle:

30*195 mm, Omni International, Kennesaw, GA, USA) at solid-to-liquid ratio (1: 50,

w/v) for 5 min was the optimal condition for extraction. Therefore, ground catfish skins

were mixed with hydrochloric acid (store in 4°C for 6 hours in advance) at pH 2.3, 2.4

and 2.5 with solid-to-liquid ratios varying from 1:15 to 1:50 (W/V). The mixture was

then homogenized at 7000 rpm for 5 s and stop 5s, the cycle was repeated until 5 min was

achieved, the mixture was kept in a foam box, which was pre-filled with ice cube for the

homogenization processing. After homogenization, the mixture was stirred for 1 hr in

4°C. After the desire time was achieved, the mixture was centrifuged at the same

conditions in as that in Section 3.3.4, and the following procedure was the same as

described in section of 3.3.4.

3.3.6 Extraction of collagen with pepsin and homogenization aided (PHSC)

method

Catfish skins and hydrochloric acid (pH 2.4) were mixed with solid-to-liquid

ratios varying from 1:5 to 1:20, and pepsin concentrations from 0.118 to 23.6 KU/g skin

(wet basis). The mixture was homogenized at 7000 rpm for 5 s and stop 5s, the cycle was

repeated until 5 min was achieved. The mixture was centrifuged under the same

conditions as described in Section 3.3.4, the following procedures were the same as that

described in Section 3.3.4.

44

3.3.7 Kinetic analysis of collagen extraction

All extraction curves could be described by the model developed by Peleg (Peleg,

1988):

𝐶(𝑡) = 𝐶(0) +𝑡

𝐾1+𝐾2·𝑡 (3.2)

Where C(t) (%) is the hydroxyproline recovery rate at time t, t is the extraction time

(hours), C0 (%) is the hydroxyproline recovery rate at time t = 0, K1 is Peleg’s rate

constant and K2 is Peleg’s capacity constant. C0 in all experimental was zero, the above

equation was rearranged in the following form:

𝐶(𝑡) =𝑡

𝐾1+𝐾2·𝑡 (3.3)

The Peleg’s rate constant K1 relates to extraction rate (B0) at the very beginning ：

𝐵0 =1

𝐾1 (3.4)

The Peleg’s capacity constant K2 relates to maximum of hydroxyproline recovery rate.

When t → ∞, the following equation describes the relations between hydroxyproline

recovery rate and K2 constant:

C|t → ∞ =1

𝐾2 (3.5)

Therefore, equation (1) can be transformed to a linear relationship in the following form:

𝑡

𝐶(𝑡)−𝐶(0)= 𝐾1 + 𝐾2𝑡 (3.6)

3.3.8 Sodium-dodecyl-sulfate polyacrylamide gel electrophoresis (SDS-PAGE)

SDS-PAGE was carried out according to the method of Laemmli (1970).

Collagen samples were dissolved in 0.02 M sodium phosphate (pH 7.2) containing 0.5 M

urea. Electrophoresis was performed on a 6% resolving gel with a 4% stacking gel on the

45

top layer. Proteins were stained with 0.1% Commassie Brilliant Blue R-250 dissolved in

water, methanol and acetic acid (9:9:2, v/v/v) for 30 min, then destained using a solution

containing water, methanol and acetic acid (17:1:2, v/v/v). For the quantification of α and

β chains, gels were scanned and analyzed by a Molecular Imager (Bio-Rad Chemidoc™

XRS+, Hercules, CA, USA) equipped with Image Lab™Analysis Software (version 5.2).

3.3.9 Effect of pH on solubility

Collagens extracted with different conditions were dissolved in 0.5 M acetic acid

to obtain a final concentration of 3 mg/mL. The pH of protein solution was adjusted with

2 mol/L HCl or 2 mol/L NaOH to 1 to 10. The solution was centrifuged at 10,000 ɡ for

20 min at 4°C. Protein content in the supernatant was determined by the method of

Bradford (1976).

3.3.10 Effect of NaCl on solubility

Collagens extracted with different conditions were dissolved in 0.5 M acetic acid

to obtain a final concentration of 6 mg/mL. Five milliliters of protein solution were

mixed with 5 mL of 0.5 M acetic acid containing a series of the concentrations of NaCl to

make a final concentration of 0%, 1%, 2%, 3%, 4%, 5% and 6%, respectively. Protein

solutions were centrifuged at 10,000 ɡ for 20 min at 4°C. The protein concentration in the

supernatant was determined by the method of Bradford (1976).

3.3.11 Zeta (ζ) potential

Collagens extracted by the above three methods with optimal conditions were

used. Zeta potential of collagen samples was determined by a Zeta potential analyzer

(Zetasizer Nano ZS90, Malvern Instr., UK). Collagen extracted with different conditions

46

were dissolved in 0.1 M acetic acid to make a final concentration of 1 mg/mL. One

milliliter of collagen solution was transferred to a capillary cell, and the pH of the

collagen solutions was adjusted to 2-6, using 1 M nitric acid or 1 M KOH. The Zeta

potential of each solution was measured in triplicate.

3.3.12 Circular dichroism (CD)

Collagen extracted by the above three methods with optimal conditions was used.

The prepared collagen samples were dissolved in 0.1 M acetic acid and diluted to a

concentration of 0.2 mg/mL with 0.1 M acetic acid solution. CD was measured using a

Jasco 810 spectropolarimeter (Jasco International Co., Easton, MD) with a Peltier

jacketed software-controlled multiple cuvette holder. The CD spectra of the samples were

measured using a 0.1 cm path length quartz cell for far-UV measurement. The spectrum

was recorded from 190-250 nm in 0.1 nm steps with response times of 1 s. Five scans

were averaged for each sample.

3.3.13 Differential scanning calorimetry (DSC)

Collagen extracted by the above three methods with optimal conditions was used.

Differential scanning calorimetry of collagen was run according to the method reported

previously with some modifications (Singh, Benjakul, Maqsood, & Kishimura, 2011).

Collagen was rehydrated by 0.05 M acetic acid at a solid to liquid ratio of 1:40 (w/v). The

mixtures were allowed to stand at 4°C for 2 days. DSC was carried out using a

differential scanning calorimetry (Perkin Elmer, Model DSC8000, Norwalk, CA, USA).

Forty µL of samples were accurately pipetted into an aluminum pan and sealed, and an

empty pan was used as reference. The sample was scanned at 1°C/min over the range of

47

20-50°C in a nitrogen atmosphere.

3.3.14 Gel strength

The bloom strength of collagen extracted by the above three methods with

optimal conditions was determined according to the British standard BS 757 (BSI, 1975).

The gels were formed in a standard bloom bottle by mixing 7.5 g of dry collagen samples

with 105 mL distilled water (6.67%, W/V), and stir for 1 hr, then brought to 60°C in a

water bath for 30 min, then were cooled down in a 4°C refrigerator for 12-16 hr to form

gels. Bloom gel strength was determined using a TA.XTplus Texture Analyzer (Stable

Microsystems, Godalming, Surrey, UK), equipped with a 0.5 inch diameter plunger with

cross-head speed of 0.5 mm/s. The standard glass bloom bottle was placed on the center

of the stage. The maximum force (ɡ) was determined when the plunger penetrated into 4

mm of the gel. Three replications were averaged for each collagen sample.

3.3.15 Statistical analysis

Experiments were performed based on a completely randomized design.

Extraction was performed in triplicate. Data were analyzed by ANOVA using 2015 SAS

(version 9.3, SAS Inc., Cary, NC, USA). Duncan’s multiple range test was carried out to

determine any significant differences among various extraction conditions using p=0.05.

3.4 Results and discussion

3.4.1 Proximate analysis of catfish skin

Several studies had reported the protein content in fish skin, Nile perch skin

contained 21.6% protein (Muyonga, Cole, & Duodu, 2004a) and bigeye snapper skin

possessed 32% of protein (Kittiphattanabawon, Benjakul, Visessanguan, Nagai, &

48

Tanaka, 2005) on a dry basis. Protein content in catfish skin was determined in this study

by the Kjeldahl method and a conversion factor of 5.4 was used. The protein content in

catfish skin was 28.88% based on the wet weight basis, and 81.21% on the dry weight

basis, which is comparable to the above fish species. Lipid, ash and carbohydrate

contents in catfish skin are 14.29%, 1.88% and 2.62% (dry weight basis), respectively.

Theoretically, the collagen yield should not be higher than the total protein content in

catfish skin. Taking the energy, solvent and time input into consideration, an extraction

method with lower energy input and high yield is expected by the food industries.

3.4.2 Collagen extracted with acids and proximate analysis

In most studies, 0.1 M acetic acid (Liu, et al., 2015; Nagai & Suzuki, 2000) or 0.5

M acetic acid (Chen, Li, Yi, Xu, Gao, & Hong, 2016; Kittiphattanabawon, Benjakul,

Visessanguan, Nagai, & Tanaka, 2005; Muyonga, Cole, & Duodu, 2004a; Nagai, Araki,

& Suzuki, 2002) were used to extract skin collagen. In the current study, different acids,

liquid-to-solid ratios and pH’s were used to extract catfish skin collagen to search for the

extraction conditions for optimal yield and quality. The yield of collagen extracted with

different acids was expressed as hydroxyproline recovery rate (%), and results are shown

in Figure 3.1 (A). The hydroxyproline recovery rate from the minced skin extracted with

pH 2.4 hydrochloric acid was the highest (42.36%) followed by extraction with pH 2.7

acetic acid (39.45%). Lactic acid and citric acid are less effective in terms of

hydroxyproline recovery rates. And this observation was in conflict with the report,

which claimed that HCl was the least effective solvent for collagen extraction from cod

skin (Skierka & Sadowska, 2007). The reason might be that different concentrations

(0.15 M HCl) of acids were used to extract collagen from cod skin, and that the pH of the

49

mixture was not maintained the same as the beginning pH, which tended to change over

the time of extraction.

Collagen was successfully isolated from several fish species with the yields

ranging from 2% to 51.4% (Chen, Li, Yi, Xu, Gao, & Hong, 2016; Liu, Li, & Guo, 2007;

Nagai, Araki, & Suzuki, 2002; Kittiphattanabawon, Benjakul, Visessanguan, Nagai, &

Tanaka, 2005; Ogawa, Moody, Portier, Bell, Schexnayder, & Losso, 2003; Duan, Zhang,

Du, Yao, & Konno, 2009; Nagai, Yamashita, Taniguchi, Kanamori, & Suzuki, 2001;

Nagai & Suzuki, 2000). However, due to the impurity of the extracted collagen, some of

the reported yields might be higher than the true value. Even though extractions for

collagen from fish skins have been reported in several studies, most of them used a single

set of extraction conditions. A systematic approach for collagen extraction, particularly

for catfish collagen extractions was lacking, and therefore making comparison between

our and reported studies difficult.

3.4.3 Collagen extraction-aided by homogenization with or without pepsin

Since pH 2.4 HCl gave the highest hydroxyproline recovery rate from

experiments conducted in Section 3.4.2, therefore, the pH interval was further decreased

to investigate the optimal pH in this step. The results of hydroxyproline recovery rate of

collagen extracted with homogenization-aided method are shown in Figure 3.1 (B). The

hydroxyproline recovery rate from minced skin extracted with pH 2.4 hydrochloric acid

at the solid-to-liquid ratio of 1:50 (W/V) was the highest (60.38%) but with no significant

differences from that extracted by pH 2.4 hydrochloric acid at the solid-to-liquid ratio of

1:40 (w/v) (57.68%). Taking the solvent input into consideration, extraction conditions of

pH 2.4 hydrochloric acid and the solid-to-liquid ratio of 1:40 (w/v) were optimal

50

conditions for scaling up in the food industries. Studies had reported collagen extracted

with homogenization-aided method. Collagen from Plaice skin was extracted with

homogenization-aided method in 0.2 and 0.4 M NaCl solution. Unfortunately, no yield

data were provided in this study (Montero, Alvarez, Marti, & Borderias, 1995). An

extrusion-hydro extraction method had been used to extract collagen from tilapia’s scale

with an yield of 16.9% (wet basis) (Huang, Kuo, Wu, & Tsai, 2016).

Peptides in the telopeptide region can be cleaved by pepsin, with the limited

concentrations of pepsin, the cross-linked molecules at the telopeptide region can be

cleaved by pepsin but without altering the triple helix structure of collagen (Liu, Li,

Miao, & Wu, 2009). Therefore, the extraction of collagen partly cleaved by pepsin gave a

higher yield (Chen, et al., 2016; Nagai & Suzuki, 2002; Nagai, Yamashita, Taniguchi,

Kanamori, & Suzuki, 2001; Tang, Chen, Su, Weng, Osako, & Tanaka, 2015; Zhang, Liu,

Li, Shi, Miao, & Wu, 2007). Although pepsin had been used widely for the collagen

extraction, most researchers reported the additions of pepsin on a weight basis instead of

enzyme activity unit basis (Chen, et al., 2016; Liu, Li, & Guo, 2007; Nalinanon,

Benjakul, Visessanguan, & Kishimura, 2007; Veeruraj, Arumugam, Ajithkumar, &

Balasubramanian, 2015; Zhang, Liu, Li, Shi, Miao, & Wu, 2007). However, only one

report used enzyme activities for calculating pepsin concentration (Nalinanon, Benjakul,

Visessanguan, & Kishimura, 2007) for collagen extraction from the Bigeye snapper skin.

Results of hydroxyproline recovery rates of collagen extracted with different

concentrations of pepsin and liquid-to-solid ratios are shown in Figure 3.1 (C). The

hydroxyproline recovery rate from minced skin extracted with pH 2.4 hydrochloric acid

at the solid-to-liquid ratio of 1:30 (W/V) with pepsin content 23.6 KU/g was the highest

51

(59.03%) but with no significant differences from that extracted by pH 2.4 hydrochloric

acid at the solid-to-liquid ratio of 1:20 (W/V) and with pepsin content 5.90 KU/g

(56.77%). In most collagen extraction studies with pepsin addition, longer time periods

(24-48 hrs) were used to extract collagen at a low pH range (Nalinanon, Benjakul,

Visessanguan, & Kishimura, 2007; Pal, Nidheesh, & Suresh, 2015). However, we are the

first to shorten the extraction time to less than 1 hr without compromising collagen yield.

Comparing the extraction method aided by homogenization with or without pepsin, the

highest hydroxyproline recovery rate was not significantly different (64.19% and

61.80%, respectively); however, extraction aided with pepsin 55 min quicker and 2.5

times of less solvent were used than that of extraction without pepsin. Therefore,

homogenization-aided extraction with pepsin addition could be more practical for the

food industries in terms of water saving, which is important in waste treatment in the

industry to influence the overall cost of production.

52

Figure 3.1 Hydroxyproline recovery rate of collagen extracted with different acids.

(A): Hydroxyproline recovery rate of collagen extracted with different acids; (B):

Hydroxyproline recovery rate of collagen extracted by HCl with homogenization aided

method; (C): Hydroxyproline recovery rate of collagen extracted by HCl with pepsin

aided method.

0.00%

5.00%

10.00%

15.00%

20.00%

25.00%

30.00%

35.00%

40.00%

45.00%

1.5 1.8 2.1 2.4 2.7 3 3.3

Hyd

rox

ypro

lin

e re

cover

y ra

te (

%)

pH

Citric acid

HCl

Acetic acid

Lactic acid

A

0.00%

10.00%

20.00%

30.00%

40.00%

50.00%

60.00%

70.00%

10 15 20 25 30 35 40 45 50 55

Hyd

rox

ypro

lin

e re

cover

y ra

te (

%)

Liquid to solid ratio (V/W)

2.3 HCl

2.4 HCl

2.5 HCl

B

53

Figure 3.1 (continued)

Place all detailed caption, notes, reference, legend information, etc here

3.4.4 Kinetic analysis of collagen extraction

Kinetic analysis was carried out for collagen extraction at different extraction

conditions (different acids, pH, liquid to solid ratio, particle size and enzyme

concentration). The Peleg’s model was used for the kinetic analysis. Since most studies

used 0.5 M acetic acid (pH around 2.6) to extract skin collagen, therefore, we compared

four different acids to extract catfish skin collagen at this pH. The effect of different acids

(pH 2.6) on collagen extraction is shown in Figure 3.2 (A). As expected, hydroxyproline

recovery rate showed an overall increasing pattern with time extended. The extraction

curves indicated that hydrochloric acid gave the highest hydroxyproline recovery rate and

the lowest in the case of citric acid extraction. These four types of acids exhibited a sharp

increase in hydroxyproline recovery rates in the first 12 hrs, and then reached a plateau-

like pattern. Figure 3.2 (B-F) showed the effect of pH of HCl, solid-to-liquid ratio

0.00%

10.00%

20.00%

30.00%

40.00%

50.00%

60.00%

70.00%

0.00 5.00 10.00 15.00 20.00 25.00

Hyd

roxyp

roli

ne

reco

ver

y ra

te (

%)

Enzyme concentration (KU/g of skins)

1:5 (V/W)

1:10 (V/W)

1:15 (V/W)

1:20 (V/W)

1:30 (V/W)

C

54

(without homogenization), particle size of catfish skin, solid-to-liquid ratio (with

homogenization aided) and pepsin concentration (with homogenization aided) on

hydroxyproline recovery rates. Figure 3.2 (B) exhibited the influence of pH on the

hydroxyproline recovery rate, the highest hydroxyproline recovery rate occurred at pH

2.4 followed by pH 2.1. A significantly low hydroxyproline recovery rate (<10%) was

observed at pH 3.0, and the increasing patterns were similar with that in Figure 3.2 (B).

Therefore, HCl with pH 2.4 was used for the following experiments. Figure 3.2 (C)

showed the effect of different solid-to-liquid ratios on the hydroxyproline recovery rate,

and the highest hydroxyproline recovery rate was obtained at the highest solid-to-liquid

ratio (1:50, w/v). Figure 3.2 (D) showed the influence of particle size on hydroxyproline

recovery rate, the highest hydroxyproline recovery rate was obtained at particle size less

than 0.5 mm, and the lowest hydroxyproline recovery rate was for the extraction of the

whole skins. However, no other papers studied particle size effect, this makes comparing

with other collagen extraction yield difficult. Hydroxyproline recovery rate exhibited a

sharp increasing in the first 48 hrs, and approaching the equilibrium hydroxyproline

recovery rate. Our objective was to develop a method with less extraction time and

solvent input, that would be useful for scale-up commercialization; therefore,

homogenization-aided method was applied. Figure 3.2 (E) showed the effect of different

solid-to-liquid ratios on hydroxyproline recovery rates when homogenization was

incorporated in the extraction process. The highest hydroxyproline recovery rate was

obtained at solid-to-liquid ratio of 1:50 (w/v), but not significantly (P>0.05) higher than

1:40 (w/v). However, the solvent input at 1:40 ratio was still high for the scale-up

extraction in the food industries. Therefore, enzyme-aided extraction method was applied

55

as described in Section 3.4.3. Figure 3.2 (F) shows that the effect of pepsin concentration

on hydroxyproline recovery rate when the extraction aided by homogenization during

extraction. The highest hydroxyproline recovery rate was obtained at pepsin

concentration of 23.6 KU/g, but not significantly higher than that of pepsin concentration

at 5.9 KU/g. There are no reported studies to be compared with our results since our

study is the first to conduct kinetic analysis on fish skin collagen extraction. Comparing

to other fish skin collagen extraction studies, 51.4% of collagen (based on dry weight)

were extracted from Japanese sea-bass with a process length of a total of 5 days.

Unfortunately, no solid-to-liquid ratios were reported (Nagai & Suzuki, 2000).

Approximately 27.2% of collagen was extracted from tilapia (based on dry weight) with

an extraction time of over 24 hours, however, no solid-to-liquid ratios were provided

(Chen, Li, Yi, Xu, Gao, & Hong, 2016). Forty-six percent of collagen were extracted

from carp skin with an extraction time of over 24 hrs and 40 volumes of liquid input (Liu,

et al., 2015). The Peleg’s extraction constants are shown in Table 3.1. K1 relates to

extraction rate at the very beginning, and Peleg capacity constant K2 relates to the

maximum yield of the extraction. As shown in Table 3.1, the lowest K1 and K2 were

found in pepsin-aided extraction with enzyme activity of 23.6 KU/g. The values of

extraction rate constant (K1) and constant of extraction extent (K2) exhibited a tendency

of decreasing with the increase of solid to liquid ratio and pepsin concentration. The

lower the extraction rate constant and constant of extraction extent suggested a higher

yield might be attained with less extraction time.

56

Fig

ure

3.2

H

ydro

xypro

line

reco

ver

y r

ate

of

coll

agen e

xtr

acte

d w

ith d

iffe

rent

acid

s at

pH

2.6

.

(A):

Hydro

xypro

line

reco

ver

y r

ate

of

coll

agen e

xtr

acte

d w

ith d

iffe

rent

acid

s at

pH

2.6

; (B

): H

ydro

xypro

line

reco

ver

y r

ate

of

coll

agen e

xtr

acte

d w

ith H

Cl at

dif

fere

nt

pH

.

0.0

0%

5.0

0%

10.0

0%

15.0

0%

20.0

0%

25.0

0%

012

24

36

48

60

72

84

Hydroxyproline recovery rate (%)

Tim

e (h

ours

)

Cit

ric

acid

HC

l

Lac

tic

acid

Ace

tic

acid

A

0.0

0%

5.0

0%

10.0

0%

15.0

0%

20.0

0%

25.0

0%

30.0

0%

012

24

36

48

60

72

84

Hydroxyproline recovery rate (%)

Tim

e (h

ours

)

pH

1.8

pH

2.1

pH

2.4

pH

2.7

pH

3.0

B

57

Fig

ure

3.2

(co

nti

nued

)

(C):

Hydro

xypro

line

reco

ver

y r

ate

of

coll

agen e

xtr

acte

d b

y p

H 2

.4 H

Cl w

ith d

iffe

rent

soli

d t

o l

iqu

id r

atio

(W

/V);

(D

):

Hydro

xypro

line

reco

ver

y r

ate

of

coll

agen e

xtr

acte

d b

y p

H 2

.4 H

Cl (W

/V=

1:5

0)

wit

h d

iffe

rent

skin

par

ticle

siz

e.

0.0

0%

5.0

0%

10.0

0%

15.0

0%

20.0

0%

25.0

0%

30.0

0%

35.0

0%

012

24

36

48

60

72

84

Hydroxyprolinerecovery rate (%)

Tim

e (h

ours

)

1:1

0

1:2

0

1:3

0

1:4

0

1:5

0

C

0.0

0%

5.0

0%

10.0

0%

15.0

0%

20.0

0%

25.0

0%

30.0

0%

012

24

36

48

60

72

84

Hydroxyprolinerecovery rate (%)

Tim

e (h

ours

)

Whole

skin

s

2-4

mm

1-2

mm

0.5

-1 m

m

<0.5

mm

D

58

Fig

ure

3.2

(co

nti

nued

)

(E):

Hydro

xypro

line

reco

ver

y r

ate

of

coll

agen e

xtr

acte

d b

y p

H 2

.4 H

Cl w

ith d

iffe

rent

soli

d t

o l

iqu

id (

W/V

) ra

tio

wit

h

ho

mo

gen

izat

ion a

ided

; (F

): H

ydro

xypro

line

reco

very

rat

e o

f co

llag

en

extr

acte

d b

y p

H 2

.4 H

Cl (W

/V=

1:5

0)

wit

h d

iffe

rent

pep

sin

conce

ntr

atio

n.

0.0

0%

10.0

0%

20.0

0%

30.0

0%

40.0

0%

50.0

0%

60.0

0%

70.0

0%

02

46

8

Hydroxyprolinerecovery rate (%)

Tim

e (h

ours

)

1:1

0

1:2

0

1:3

0

1:4

0

1:5

0

E

0.0

0%

10

.00

%

20

.00

%

30

.00

%

40

.00

%

50

.00

%

60

.00

%

70

.00

%

02

46

8

Hydroxyprolinerecovery rate (%)

Tim

e (h

ours

)

0.1

18 K

U/g

1.1

18 K

U/g

5.9

0 K

U/g

11.8

KU

/g

23.6

KU

/g

F

59

Table 3.1 Kinetic constant and R2 for hydroxyproline recovery rate under different

extraction conditions.

K1 K2 R2

Acid

Citric acid 30.47 7.39 0.989

Lactic acid 20.49 6.74 0.994

Acetic acid 13.91 5.99 0.997

Hydrochloric acid 13.23 5.46 0.996

pH

pH 1.8 38.86 8.16 0.984

pH 2.1 18.69 5.00 0.990

pH 2.4 14.39 3.57 0.987

pH 2.7 26.78 6.78 0.990

pH 3.0 110.03 3671 0.994

Ratio (without homogenization)

1:10 53.59 13.39 0.990

1:20 31.74 7.26 0.988

1:30 20.25 5.29 0.991

1:40 19.33 4.37 0.987

1:50 13.81 3.53 0.987

Particle size

Whole skins 31.39 8.77 0.992

2-4 mm 24.47 7.23 0.992

1-2 mm 19.65 6.30 0.994

0.5-1 mm 18.64 4.52 0.989

<1 mm 15.54 3.73 0.989

Ratio (with homogenization)

1:10 2.49 12.70 0.997

1:20 0.57 3.34 0.998

1:30 0.23 2.12 0.999

1:40 0.15 1.70 0.999

1:50 0.11 1.63 0.999

Pepsin concentration

0.118 KU/g 0.19 1.81 0.999

1.18 KU/g 0.14 1.68 0.999

5.90 KU/g 0.14 1.59 0.999

11.8 KU/g 0.11 1.57 0.999

23.6 KU/g 0.08 1.55 0.999

60

3.4.5 SDS-PAGE of collagen extracted with different conditions

Figures 3.3 and 3.4 show the SDS-PAGE patterns of collagen extracted with

different conditions. All collagen products extracted with different conditions had two

different α chains (α1 and α2) and their cross-linked chains (dimer is referred to as β

chain, trimer is referred to as γ chain). The molecular mass of α1 chain is 123 KDa and

113 KDa for α2 chain. The α1 chain might contain a α3 chain that has a molecular mass

close to α1 chain, resulting a indistinguishable band on a SDS-PAGE gel (Kimura, Zhu,

Matsui, Shijoh, & Takamizawa, 1988). And the molecular mass of the subunit was 226

KDa for β chain and 338 KDa for γ chain, respectively. These values were similar to the

molecular mass of black drum and sheepshead seabream skin collagen (Ogawa, Moody,

Portier, Bell, Schexnayder, & Losso, 2003), carp skin collagen (Duan, Zhang, Du, Yao,

& Konno, 2009) and tilapia skin collagen (Chen, Li, Yi, Xu, Gao, & Hong, 2016).

However, extracted collagens described in the above literature were not from the Channel

catfish species. Channel catfish skin collagen was successfully isolated but with the mass

of the subunits had not been reported (Liu, Li, & Guo, 2007). In addition, skin collagen

of striped catfish was isolated and characterized before, however, no molecular mass of

each subunit was provided (Singh, Benjakul, Maqsood, & Kishimura, 2011). The

existence of two different α chains confirmed that the major collagen from catfish skins

was of the type-I collagen (Sato, 1993). Band intensity ratio of β/(α1+ α2) chains in acid-

solubilized collagen (ASC), homogenization-solubilized collagen (HSC) and pepsin-

homogenization solubilized collagen (PHSC) were 1.94, 1.72 and 0.41, respectively. The

quantity of β chain was higher in acid-extracted collagen than that in pepsin-aided

extracted collagen, in which 55.5% of the β chain of the pepsin-aided extracted collagen

61

degraded into α1 and α2 chains. The reason might be that some β chains were converted to

α chains by pepsin, this observation was similar to the study of skin collagen from

Channel catfish, black drum and sheepshead seabream (Ogawa, Moody, Portier, Bell,

Schexnayder, & Losso, 2003). As mentioned above, pepsin removes the cross-linked

containing telopeptide, and one β chain is converted to two α chains (Sato, Ebihara,

Adachi, Kawashima, Hattori, & Irie, 2000). Since no studies had used homogenization-

aided method to extract fish skin collagen, therefore, our studies presented the first

homogenization effect on catfish collagen extraction yield and quantity.

62

Fig

ure

3.3

S

DS

-PA

GE

gel

s o

f co

llag

ens

extr

acte

d w

ith v

ario

us

acid

extr

acti

ons.

Left

fig

ure

: L

ane

1:

HC

l, p

H 3

extr

acte

d c

oll

agen;

Lane

2:

HC

l, p

H 2

.7 e

xtr

acte

d c

oll

agen;

Lane

3:

HC

l, p

H 2

.4 e

xtr

acte

d c

oll

agen;

Lane

4:

HC

l, p

H 2

.1 e

xtr

acte

d c

oll

agen;

Lane

5:

HC

l, p

H 1

.8 e

xtr

acte

d c

oll

age;

Lane

6:

Ace

tic

acid

, pH

3 e

xtr

acte

d c

oll

agen;

Lane

7:

Ace

tic

acid

, pH

2.7

extr

acte

d c

oll

agen;

Lane

8:

Ace

tic

acid

, pH

2.4

extr

acte

d c

oll

agen;

Lane

9:

Acet

ic a

cid

, pH

2.1

extr

acte

d

coll

agen;

Lane

10:

Ace

tic

acid

, pH

1.8

extr

acte

d c

oll

agen.

Rig

ht

figure

: L

ane

1:

Lac

tic

acid

, pH

3 e

xtr

acte

d c

oll

agen;

Lane

2:

Lac

tic

acid

, pH

2.7

extr

acte

d c

oll

agen;

Lane

3:

Lac

tic

acid

, pH

2.4

extr

acte

d c

oll

agen;

Lane

4:

Lac

tic a

cid

, pH

2.1

extr

acte

d

coll

agen;

Lane

5:

Lac

tic

acid

, pH

1.8

extr

acte

d c

ollag

en;

Lane

6:

Cit

ric

acid

, pH

3 e

xtr

acte

d c

oll

agen;

Lan

e 7:

Cit

ric

acid

, pH

2.7

extr

acte

d c

oll

agen;

Lane

8:

Cit

ric a

cid

, pH

2.4

extr

acte

d c

oll

agen;

Lane

9:

Cit

ric

acid

, pH

2.1

extr

acte

d c

oll

agen;

Lane

10:

Cit

ric

acid

, pH

1.8

extr

acte

d c

oll

agen.

63

Fig

ure

3.4

S

DS

-PA

GE

gel

of

coll

agens

extr

acte

d w

ith d

iffe

rent

conce

ntr

atio

ns

of

pep

sin.

Lane

1:

HC

l at

pH

2.4

, W

/V =

1:5

, ho

mo

gen

izat

ion,

pep

sin c

once

ntr

atio

n 1

18U

/g;

Lane

2:

HC

l at

pH

2.4

, W

/V =

1:5

,

ho

mo

gen

izat

ion,

pep

sin c

oncentr

atio

n 1

180U

/g;

Lan

e 3:

HC

l at

pH

2.4

, W

/V =

1:5

, ho

mo

geniz

atio

n,

pep

sin c

oncentr

atio

n

5900U

/g);

Lane

4:

HC

l at

pH

2.4

, W

/V =

1:5

, ho

mo

gen

izat

ion,

pep

sin c

once

ntr

atio

n 1

1800U

/g;

Lane 5

: H

Cl at

pH

2.4

, W

/V =

1:5

,

ho

mo

gen

izat

ion,

pep

sin c

oncentr

atio

n 2

3600U

/g;

Lane

6:

HC

l at

pH

2.4

, W

/V =

1:2

0,

ho

mo

gen

izat

ion,

pep

sin c

oncentr

atio

n

118U

/g;

Lane

7:

HC

l at

pH

2.4

, W

/V =

1:2

0,

ho

mo

gen

izat

ion,

pep

sin c

once

ntr

atio

n 1

180U

/g;

Lane

8:

HC

l at

pH

2.4

, W

/V =

1:2

0,

ho

mo

gen

izat

ion,

pep

sin c

oncentr

atio

n 5

900U

/g;

Lan

e 9:

HC

l at

pH

2.4

, W

/V =

1:2

0,

ho

mo

geniz

atio

n,

pep

sin c

once

ntr

atio

n

11800U

/g;

Lane

10:

HC

l at

pH

2.4

, W

/V =

1:2

0, hom

ogeniz

atio

n,

pep

sin c

oncentr

atio

n 2

3600U

/g.

64

3.4.6 Effect of pH and NaCl on collagen solubility

3.4.6.1 Effect of pH on collagen solubility

Collagen extracted with pH 2.4 HCl for 48 hrs (ASC), homogenized with pH 2.4

HCl for 5 min (HSC) with (PHSC, pepsin concentration 11.8 KU/g) or without pepsin

were used for solubility characterization. The effect of pH on the solubility of collagen

extracted by different methods is shown in Figure 3.5 (A). ASC, HSC, and PHSC showed

the highest solubility at pH 2 (p < 0.05). All the collagens were solubilized in the low pH

range (1 to 4), and a sharp solubility decrease was observed when pH was higher than 5

(p < 0.05). When the pH was not equal to pI, the net charge of the protein molecules was

larger than zero, therefore, the solubility of the protein was increased by the repulsion

forces between chains. In contrast, when the pH was equal or close to the pI, the total net

charge of the protein molecules approached zero and resulted in precipitation. It had been

reported that the pI of collagen is varied from 6 to 9 (Foegeding, Lanier, & Hultin, 1996).

The minimum solubility of collagen is around pH 6 in the current study, and this

observation was similar with the solubility of collagen from brownstripe red snapper skin

(Jongjareonrak, Benjakul, Visessanguan, Nagai, & Tanaka, 2005). Acid-extracted

collagen exhibited higher solubility than the other two collagens at pH 1 to 4 but with no

significant differences (p < 0.05). However, collagen extracted with pepsin from bigeye

snapper had been reported to have the highest solubility at pH 5 (Nalinanon, Benjakul,

Visessanguan, & Kishimura, 2007), which was closer to the isoelectric point (6 to 9).

3.4.6.2 Effect of NaCl on collagen solubility

Solubility of ASC, HSC and PHSC in 0.5 M acetic acid remained consistent at

65

NaCl concentrations of 0-2% (p > 0.05), and the results are shown in Figure 3.5 (B). A

slight decrease in solubility was observed at 3% NaCl. A sharp decrease was observed

when the concentration of NaCl increased to 4% and above (p < 0.05). Solubility of

collagen extracted from the skin of bigeye snapper, striped catfish, tilapia and brown

stripe red snapper (Jongjareonrak, Benjakul, Visessanguan, Nagai, & Tanaka, 2005;

Kittiphattanabawon, Benjakul, Visessanguan, Nagai, & Tanaka, 2005; P Montero,

Gómez-Guillén, & Borderıas, 1999; Singh, Benjakul, Maqsood, & Kishimura, 2011) also

decreased with the increasing concentration of sodium chloride. When higher

concentration of NaCl (> 3%) was presented, an increasing ionic strength might lead to a

reduction in protein solubility by increasing the interactions between protein chains

(salting out phenomenon). Thus, solubility of protein might be decreased via increasing

hydrophobic interaction and aggregation, in which salt competes with protein molecules

for water. Collagen extracted by the homogenization method with pepsin addition

showed a higher solubility than that isolated from the other two methods, and this

observation was in accordance with the collagen extracted with acid and pepsin from

striped catfish skin (Singh, Benjakul, Maqsood, & Kishimura, 2011). Greater solubility of

PHSC was due to the action of pepsin in altering the collagen structure and altered the

compositions of α, β and γ chains as discussed in Section 3.4.5. Therefore, collagen

extracted by different methods might have slightly different molecular properties.

Comparing Figure 3.3 and Figure 3.4, pepsin-added extraction produced a high band

intensity of α1 and α2 chains, which were derived from the β chain in non-pepsin

extracted collagen (Figure 3.3).

66

Figure 3.5 Effect of pH and NaCl on collagen solubility.

(A): Effect of pH on collagen solubility; (B): Effect of NaCl on collagen solubility.

3.4.7 Zeta (ζ) potential

Zeta potential is used as a factor for determining the physical properties of

colloidal dispersions. It is the e- potential difference between dispersion medium and the

0.00%

10.00%

20.00%

30.00%

40.00%

50.00%

60.00%

70.00%

80.00%

90.00%

100.00%

0 2 4 6 8 10 12

Rel

ativ

e co

llage

n s

olu

bili

ty (%

)

pH

ASC

HSC

PHSC

0.00%

10.00%

20.00%

30.00%

40.00%

50.00%

60.00%

70.00%

80.00%

90.00%

100.00%

0 1 2 3 4 5 6

Rel

ativ

e co

llage

n s

olu

bili

ty (%

)

Concentration of NaCl (% (W/V))

ASC

HSC

PHSC

A

B

67

slipping plane of the dispersed molecules/particles. A high zeta potential will give high

stability to resist aggregation. In contrast, when the zeta potential is small or close to

zero, attractive forces may be larger than repulsion forces, and the dispersion may form

aggregates. The zeta potentials of ASC, HSC and PHSC solutions at different pHs are

shown in the Figure 3.6 (A). The zero-surface net charge of ASC, HSC and PHSC was

observed at pH of 5.34, 5.73 and 5.42, respectively. The total net charge of protein would

be zero when the pH of protein close to isoelectric point (Vojdani, 1996). The results of

the effect of pH on protein solubility indicated that PHSC had the lowest solubility at pH

6, which suggested that the isoelectric point of ASC, HSC and PHSC was around 6.

Thus, the results of zeta potential and solubility test were mutual supportive. And the

differences in surface charge between ASC and HSC might be due to the differences in

acidic and basic amino acid residues, which were likely influenced by the removal of

telopeptides by high-speed homogenization. The zero surface charge of collagen

extracted by acid and pepsin from striped catfish were 4.72 and 5.43, respectively (Singh,

Benjakul, Maqsood, & Kishimura, 2011). This discrepancy in surface charge of acid and

pepsin solubilized collagens might be attributes to the different acidic and basic amino

acid residues, which was more controlled by the removal for telopeptides by pepsin

which discussed in Section 3.4.5 (Sato, Ebihara, Adachi, Kawashima, Hattori, & Irie,

2000). Collagens extracted with acid and pepsin from brown-banded bamboo shark skin

were reported to have a zero surface charge at 6.21 and 6.56, respectively

(Kittiphattanabawon, Benjakul, Visessanguan, Kishimura, & Shahidi, 2010). In this

study, PHSC showed a higher zero surface net charge (pH 5.42) than ASC (pH 5.34), and

this observation was in accordance with other studies described above.

68

3.4.8 Circular dichroism (CD)

CD spectra of collagen extracted with different methods are shown in Figure 3.6

(B). CD spectrum with positive peaks at 221 nm for both ASC and PHSC and 222 nm for

HSC were observed; and negative extreme for HSC and PHSC were observed at 199 nm,

ASC showed a negative extreme at 200 nm. Slight deviations in ellipticity were observed

among the three collagen products extracted with different methods, which indicated that

there was a minor alternation in the structures of these three collagens. The spectra

characteristics in the three collagens are typical of the collagen triple helix structure

(Engel, 1987), which suggested that the triple helical structure of collagen was not

destroyed during extraction processing. CD spectra of collagen extracted from Pagrus

major, oreochromis niloticas, Labeo rihita, Catla scale, eel-fish skin, black drum and

sheepshead seabream skin, bone and scale (Ikoma, Kobayashi, Tanaka, Walsh, & Mann,

2003; Ogawa, Moody, Portier, Bell, Schexnayder, & Losso, 2003; Ogawa, Portier,

Moody, Bell, Schexnayder, & Losso, 2004; Pati, Adhikari, & Dhara, 2010; Veeruraj,

Arumugam, & Balasubramanian, 2013) had been exhibit shown to a similar pattern with

our study.

3.4.9 Differential scanning calorimetry (DSC)

DSC thermograms of collagen extracted by different methods are shown in Figure

3.6 (C). Endothermic peaks with Tmax of collagens extracted by three different method are

ranging from 35.57 to 36.12°C. The ∆H and Tmax of ASC and HSC were slightly higher

than that of PHSC. This might be caused by the different α and β chains composition

since ASC and HSC exhibited higher ratios of β/(α1+ α2) chains (discussed in section

3.5). Higher levels of cross-linkage of collagens were more likely to contribute to the

69

higher Tmax (Singh, Benjakul, Maqsood, & Kishimura, 2011). The results of DSC were

higher than previous report on the collagens extracted from Channel catfish skin (Tmax =

32.5°C) (Liu, Li, & Guo, 2007). The minor differences might be caused by the different

raw materials, the catfish used might be raised in different temperature of water. DSC

was used in the current study. However, denaturation temperature in the cited reference

was determined by monitoring the viscosity change of collagen solution under a

temperature range. Therefore, a different denaturation temperature might have occurred.

70

Fig

ure

3.6

P

hysi

cal and s

truct

ura

l char

acte

riza

tio

ns

of

extr

acte

d c

oll

agen.

(A):

Zet

a (ζ

) pote

nti

al o

f ex

trac

ted c

oll

agen.

-505

10

15

20

25

01

23

45

67

Zeta (ζ) potential (mV)

pH

AS

C

HS

C

PH

SC

71

Fig

ure

3.6

(co

nti

nued

)

(B):

CD

spec

tra

of

coll

agen e

xtr

acte

d w

ith d

iffe

rent

condit

ions.

-50

-40

-30

-20

-100

10

190

200

210

220

230

240

250

CD (mdeg)

Wav

elen

gth

(n

m)

HS

C

PH

SC

AS

C

Den

ature

d

72

Fig

ure

3.6

(co

nti

nued

)

(C):

DS

C t

her

mo

gra

ms

of

AS

C (

acid

so

lubil

ized

co

llag

en),

HS

C (

ho

mo

gen

izat

ion a

ided

so

lubil

ized

co

llag

en)

and H

PS

C (

pep

sin

and h

om

ogen

izat

ion-s

olu

bil

ized

co

llag

en)

fro

m C

han

nel

catf

ish s

kin

dis

per

sed i

n 0

.05 M

ace

tic a

cid

.

02468

10

12

14

16

20

25

30

35

40

45

50

Heat Flow (W/g)

Tem

per

atu

re (

°C)

AS

CH

SC

PH

SC

C

Tm

ax =

36.0

5°C

, Δ

H =

0.6

51

J/g

Tm

ax =

35.5

7°C

, Δ

H =

0.5

64

J/g

Tm

ax =

36.1

2°C

, Δ

H =

0.6

72

J/g

73

3.4.10 Gel strength

Gel strength is one of the most important index for collagen quality and it can be

classified into 3 levels: low (<150 ɡ), medium (150-220 ɡ) and high (220-300 ɡ)

(Johnston-Bank, 1983). The prepared collagen gels were used to test the gel strength, and

ASC, HSC and PHSC exhibited gel strength of 223.13, 220.48 and 73.59 ɡ, respectively.

This result was expected since pepsin-added extraction produced smaller α1 + α2

molecules, which would produce weaker gels. Gel strength of collagen extracted from

tilapia had been reported to be 157 to 260 ɡ at the same concentration (Huang, Kuo, Wu,

& Tsai, 2016). The discrepancy in gel strength among species might be due to the

different amino acid composition and molecular size of protein chains (Muyonga, Cole,

& Duodu, 2004b). When gel strength was measured at low temperature (< 10 °C), some

short chain peptides present in low viscosity gelatins tend to strengthen the gel (Montero

& Gómez-Guillén, 2000). In addition, the thermal shrinkage, denaturation temperature of

collagens and melting temperature of gelatins isolated from cold-water fish had been

reported to be significantly lower than collagens and gelatins from warm water fish, the

reason of the differences might be due to a lower degree of proline hydroxylation of cold-

water fish collagen (Karim & Bhat, 2009).

3.5 Conclusion

In summary, collagens were extracted and characterized from catfish skin

successfully. Hydroxyproline recovery rates from minced catfish skins extracted with pH

2.4 HCl containing 23.6 KU/g pepsin was the highest (64.19%). Kinetic analysis of

collagen extraction yield was performed, and the results showed that the lower the

74

extraction rate constant and constant of extraction extent suggested a higher yield might

be attained with less extraction time, and the kinetic results may be used as a foundation

for developing the process for future scale-up extraction in industry. Pepsin-extracted

collagen produced more smaller size chains (α chains). Gel strength of ASC, HSC was

significantly higher than that of PHSC. HCl extraction with homogenization-aided

method can be used for the industries to improve collagen yield in terms of extraction

efficiency (less solvent and time input).

75

3.6 Research diagram in proposal

Research diagram on catfish skin in proposal is shown in Figure 3.7. Future work

and limitation of the current study are discussed in Chapter VI.

Figure 3.7 Research diagram of utilization of catfish skin and characterization of

collagen extracted from catfish skin.

76

CHAPTER IV

COMPARING THE KINETIC OF THE HYDROLYSIS OF BY-PRODUCT FROM

CHANNEL CATFISH (ICTALURUS PUNCTATUS) FILLET PROCESSING BY

EIGHT PROTEASES

4.1 Abstract

Channel catfish (Ictalurus punctatus) by-products (heads and frames) were

collected from local a catfish processing plant. Papain, ficin, bromelain, neutrase,

alcalase, protamex, novo-proD and thermolysin were used for hydrolysis. Proteolytic

activities of those proteases were analyzed by using AzocollTM and Azocasein as

substrates at different hydrolysis temperatures. Proteases cost for reaching certain degree

of hydrolysis was predicted. Degree of hydrolysis (DH) of the hydrolysates and

hydrolysis kinetics were studied. Emulsifying capacities and foaming properties and

stability of selected hydrolysates were evaluated. Results indicated that thermolysin had

the highest activity (8.2×108 Azocoll U/g and 5.6×106 Azocasein U/g) at 70 °C. The

highest DH of the hydrolysates were observed in 120 min with the concentration of 80

AzU/g for all enzymes. Ficin (80 AzU/g) was the most efficient in hydrolyzing the by-

product (DH reaching 71%) in 120 min at 30°C. The hydrolysis curves fit the Peleg’s

model well with R2 higher than 0.91.

Keywords: Channel catfish, protease hydrolysis, kinetic analysis, by-product

77

4.2 Introduction

Channel catfish farming is one of the most important warm water aquaculture in

the United States and is vital to the rural economy. However, in catfish fillet processing,

approximately 60% of the whole catfish (currently the cost is about $1 per lb of the fresh

fish) is the by-product, and has been sold at 2 cents/lb to the local oil rendering plant.

Because of the steady increases in world population, global need for dietary protein

source is increasing rapidly. Catfish protein isolated from the by-product could help to

alleviate protein shortage problem, and hence help to reduce problems associated with

world hunger and malnutrition. Successful utilization of the by-product as a protein

hydrolysate may increase aquaculture profitability and sustainability.

Protein hydrolysates have been made from various fish species by bacterial

fermentation (Kristinsson & Rasco, 2000a; Sodini, Lucas, Tissier, & Corrieu, 2005) or by

enzymatic hydrolysis (Hathwar, Bijinu, Rai, & Narayan, 2011; Nilsang, Lertsiri,

Suphantharika, & Assavanig, 2005). Flavourzyme™ and Kojizyme™ have been used for

hydrolyzing tuna processing by-product (Nilsang, Lertsiri, Suphantharika, & Assavanig,

2005), alcalase and neutrase had been utilized for hydrolyzing Pacific whiting solid waste

and capelin (Benjakul & Morrissey, 1997; Shahidi, Han, & Synowiecki, 1995). Plant

proteases such as papain, ficin and bromelain have also been used for hydrolyzing fish

fillet protein (Beddows & Ardeshir, 1979; Salampessy, Phillips, Seneweera, &

Kailasapathy, 2010). Hydrolysis kinetics of fish proteins have been studied by some

research groups, and the kinetic model they used are different (Kristinsson & Rasco,

2000b; Wu, Du, Jia, & Kuang, 2016). Early research intention is to use the fish

hydrolysate for animal feed. Recently, hydrolysates of sardine and horse mackerel muscle

78

protein have been found to possess antioxidant activity (Morales-Medina, Tamm,

Guadix, Guadix, & Drusch, 2016), and hydrolysates of leatherjacket muscle protein

exhibited angiotensin-I converting enzymes (ACE) inhibition activity (Salampessy,

Reddy, Phillips, & Kailasapathy, 2017), and most of the studies have focused on

improving the degree of hydrolysis in order to improve their antioxidant activity.

However, the hydrolysis of Channel catfish by-products by using multiple proteases and

hydrolysis kinetics have not been reported. In addition, this study is the first to show the

peptide’s pattern changes during protease hydrolysis. The objective of this study was to

investigate the kinetics of the enzymatic hydrolysis of catfish by-product, and to study the

functional properties of the hydrolysates. This study provides important new engineering

information for choosing proteases and processing conditions for optimizing efficiency

and yield.

4.3 Material and methods

4.3.1 Materials

Channel catfish by-products (heads and bone frames) were collected from the

local catfish fillet processing plant (Country Select, Isola, MS), and buried under ice

during transportation and delivered to our laboratory within two hours. All materials were

stored at -80°C until use.

4.3.2 Chemicals

All chemicals were of the analytical grade. Papain, ficin, bromelain and

thermolysin used in this study were obtained from the Sigma-Aldrich Chemical Company

79

(St. Louis, MO, USA). Food grade neutraseTM, alcalaseTM, protamexTM, novo-proD®

were kindly donated by Novozyme Inc. (Franklinton, NC, USA).

4.3.3 Proteolytic activity determination by using Azocoll as substrate

A synthetic substrate AzocollTM was used to evaluate the proteolytic activity of

the proteases according to the previous method with some modifications (Kristinsson &

Rasco, 2000b). Azocoll (50 mg) was mixed with 0.1 M sodium phosphate buffer (pH 7.2)

in a 15 mL centrifuge tube to a final volume of 5 mL and incubated in a water bath at 30,

40, 50, 60 and 70°C, respectively, for 10 min. The reaction was initiated by adding

proteases (20 µL with appropriate dilution). The temperature and pH were maintained the

same as periodically throughout the hydrolysis reaction (adjust the pH every 10 min). The

tubes were incubated for 15 min in the water bath at the corresponding temperature,

inverting the tubes for every 3 min. The reaction was terminated by placing the tubes on

ice for 10 min. The mixture was centrifuged at 10000 g (4 °C) for 3 min to remove the

residue, and the absorbance of the supernatant was measured at 520 nm. The absorbance

was plotted against enzyme quantity for each assay. The results were expressed as

Azocoll unit (AzU) per gram or mL enzyme. One AzU is defined as the amount of

enzyme generating an absorption of 0.1 at 520 nm under the conditions described above.

All assays were carried out in triplicate. Proteolytic activities of catfish heads and bone

frames were determined to check if the substrates contained proteases which hydrolyzed

the proteins at significant level.

80

4.3.4 Proteolytic activity determination by using Azocasein as substrate

In order to provide the reference proteolytic activity of the eight proteases,

proteolytic activity of each protease was determined at the same temperature and pH as

described in section 4.3.3, but Azocasein was used instead of Azocoll. The procedures

were the same as the method reported before (Coêlho, Saturnino, Fernandes, Mazzola,

Silveira, & Tambourgi, 2016).

4.3.5 Enzymatic hydrolysis

Catfish heads and bone frames were grounded using a meat grinder (LEM

Products #32, West Chester, OH) for six cycles and mixed (w/w = 3:2). Distilled water

was added into the mixed by-products (w/v = 1:1) in a 2.5 L plastic bucket, and the pH of

the mixture was adjusted to 7.2. Papain, ficin, bromelain, neutraseTM, alcalaseTM,

protamexTM, novo-pro D® and thermolysin were used for hydrolysis. The reaction was

initiated by adding varying concentrations of enzymes (5, 10, 25, 40, and 80 AzU/g of

protein content in the substrate wasa determined by Kjeldahl analysis) and hydrolyzed for

10, 20, 40, 60 and 120 min at 40, 50, 60 and 70°C for thermolysin and 30, 40, 50 and

60°C for all other proteases. The pH of the mixture was maintained periodically by

adjusting the pH manually to pH 7.2. After the desired time was achieved, the reaction

was terminated by heating the mixture at 85°C for 15 min. The mixture was then

centrifuged at 10000 g for 10 min using a Sorvall rc-5b centrifuge (Dupont, Wilmington,

DE, USA) to remove the residue. The supernatant containing oil, was stored at 4°C for 2

hrs to make the top oil layer solidify for removal. The clear hydrolysates were stored at -

80°C overnight and freeze-dried. The freeze-dried powder was stored at -80°C for further

analysis.

81

4.3.6 Proximate analysis

Proximate analysis of the mixture of heads and frames was conducted. Moisture,

protein, ash and total fat content were determined by the AOAC International methods

934.01, 955.04, 942.05 and 2003.06, respectively (AOAC International, 2012). A

conversion factor of 6.25 was used for calculating the crude protein content from nitrogen

content. Protein content in hydrolysates with good functionalities were determined the

same method (method 955.04), and the protein recovery rate (%) was calculated by

dividing protein content in hydrolysates by protein content in the mixture of heads and

frames. Raw materials were placed in aluminon foil pan and dried in 105°C oven for

about 16 hours, and then stored in desiccator for further analysis. Oven dried samples

were used for protein, lipid and ash content analysis, and then calculated back to wet

weight basis.

4.3.7 Determination of the degree of hydrolysis (DH) using

trinitrobenzenesulfonic acid (TNBS) method

The degree of hydrolysis of the hydrolysates was determined according to the

method reported by Adler-Nissen (Adler-Nissen, 1979). Samples and standards were

dissolved in 1% SDS (sodium dodecyl sulfate) solution in 15 mL centrifuge tube. A

portion of 0.25 mL of the sample solution was added into a test tube, which contained 2

mL 0.02 M sodium phosphate buffer (pH 8.2). Then 2 mL of 0.1%

trinitrobenzenesulfonic acid (TNBS) was added, and the mixture incubated in the dark at

50°C for 1 hr. The reaction was terminated by acidifying with 4 mL of 0.1 N HCl. The

test tubes were allowed to stand for another 20 min to cool. The absorbance was

measured at 340 nm using Thermo Scientific Genesys 10S UV-Vis spectrophotometer

82

(Thermo Scientific, Pittsburgh, PA, USA). L-leucine (0-1 mM) was used to generate a

standard curve.

4.3.8 Kinetic analysis of hydrolysis processing

The degree of hydrolysis of the hydrolysates was plotted against hydrolysis time.

All hydrolysis curves could be described by the model established by Peleg (Peleg, 1988)

with some modifications, protease concentration was incorporated in the equation as

below.

𝐷𝐻 (𝑡) = 𝐷𝐻(0) +𝑡

𝐾1+𝐾2·1

𝐸

(4.1)

Where DH (t) (%) is the degree of hydrolysis at time t, t is the hydrolysis time (min), DH

(0) (%) is the degree of hydrolysis at time t = 0, K1 is Peleg’s rate constant and K2 is

Peleg’s capacity constant. Since DH (0) in all experimental runs was zero, the above

equation could be simplified to the following form:

𝐷𝐻 (𝑡) =𝑡

𝐾1+𝐾2·1

𝐸

(4.2)

Rate constant K1 relates to degree of hydrolysis DH (0) at the very beginning ：

DH (0) =1

𝐾1 (4.3)

The capacity constant K2 relates to the maximum of degree of hydrolysis. When t → ∞,

the following equation describes the relations between the degree of hydrolysis and K2

constant:

C|t → ∞ =1

𝐾2 (4.4)

The first equation also could be transformed into a linear relationship in the following

form:

83

𝑡

𝐷𝐻(𝑡)−𝐷𝐻(0)= 𝐾1 + 𝐾2

1

𝐸 (4.5)

4.3.9 Calculated price of proteases activity and cost to reach 15 and 20% DH

Linear relationship was obtained from the above kinetic analysis. Protease activity

and the price of proteases to reach a certain degree of hydrolysis were calculated

according to the linear relationship.

4.3.10 Sodium-dodecyl-sulfate gel electrophoresis (SDS-PAGE)

Electrophoresis was performed according to the method of Laemmli (1970). All

hydrolysates were dissolved in pH 7.2 PBS in 15 mL centrifuge tube. The protein content

was adjusted to 2 mg/mL, then mixed 1:1 (v/v) mixed with the sample buffer containing

2-mercaptoethanol. Electrophoresis was carried out on a 4-25% gradient gel (13 cm x 8

cm) in a BioRad mini-Protean chamber at 60 V for 30 min, followed by 125 V for

90 min. At the end of electrophoresis, gels were stained with 0.1% Commassie Brilliant

Blue R-250 dissolved in a solution containing water, methanol and acetic acid (9:9:2,

v/v/v) for 20 min, and destained using a solution, containing water, methanol and acetic

acid (7:2:1, v/v/v) for 6 hrs. Gels were scanned by a Molecular Imager (Bio-Rad

ChemideocTM XRS+, Hercules, CA, USA), which was equipped with Image LabTM

Analysis Software (version 5.2).

4.3.11 Emulsion capacities

Emulsifying capacities and stability were determined according the previous

method (Pearce & Kinsella, 1978; Cai, Chang & Lunde, 1996). Corn oil and 30 mL of

1% protein solution were mixed together, and the pH was adjusted to 2, 4, 6, 8 and 10,

respectively. The mixture was then homogenized using a homogenizer (Model: GLH

84

580, Spindle: 10*195 mm, Omni International, Kennesaw, GA, USA) at the speed of

10000 rpm for exactly 1 min. Fifty microliters of the emulsion solution was brought to 5

mL of 0.1% SDS solution. The absorbance of the diluted solution was measured at 500

nm with a plate reader (FlexStation 3, Molecular Devices, CA, USA). The absorbance

was measured immediately after emulsion formation and 10 min after the formation. The

emulsifying activity index (EAI) and the emulsion stability index (ESI) were calculated

as the following equation (Pearce & Kinsella, 1978). Soy protein isolate (SPI) was used

as a positive standard. The soy protein isolate was obtained according to our previous

method (Xu, Mukherjee, & Chang, 2018).

EAI (m2/g) =2 ×2.303 × A500

0.25 × protein weight (g) (4.6)

ESI (min) =A0 × Δt

ΔA (4.7)

Where ∆A = A0 – A10, and ∆t = 10 min

4.3.12 Foaming capacities

Foaming capacities and stability of hydrolysates were determined according to the

previous method (Sathe & Salunkhe, 1981). Soy protein isolate (SPI) was used as for

comparison. The soy protein isolate was obtained according to our previous method (Xu,

Mukherjee, & Chang, 2018). Twenty microliters of 0.5% of sample solution were

adjusted to pH 2, 4, 6, 8 and 10, respectively, and homogenized at a speed of 16000 rpm,

using a homogenizer (Model: GLH 580, Spindle: 10*195 mm, Omni International,

Kennesaw, GA, USA) to incorporate the air for 2 min. The whipped sample was then

transferred to a 60 mL cylinder and the total volume was read after 30 s. The foaming

capacity was calculated according to the following equation (Sathe & Salunkhe, 1981):

85

Foaming capacity (%) =𝐴−𝐵

A × 100 (4.8)

Where A is the volume after whipping (mL), B is the volume before whipping (mL).

The whipped sample was allowed to stand at 20 °C for 3 min and the volume of whipped

sample was then recorded. Foam stability was calculated as follows. The foaming

stability was calculated according to the following equation (Sathe & Salunkhe, 1981):

Foaming stability (%) =𝐴−𝐵

A × 100 (4.9)

Where A is the volume after whipping (mL), B is the volume before whipping (mL).

4.3.13 Oxygen radical absorbing capacity (ORAC) of selected hydrolysates

A plate reader (Molecular Devices, CA, USA) equipped with adjustable

fluorescence filters and incubator was used. The ORAC assay was conducted using our

published method (Prior, Wu, & Schaich, 2005; Tan, Chang & Zhang, 2017). In brief,

Hydrolysates were diluted with phosphate buffer (75 mM, pH 7.0) to the appropriate

concentration to fall within the linear range of the standard curve. Twenty microliters of

diluted hydrolysates, trolox, and blank were mixed with 200 µL pre-headed fluorescein

solution. The 96-well plate was then incubated in plate reader for 30 min at 37 oC.

Twenty microliters of 2, 2’-Azobis(2-methylpropionamidine) dihydrochloride (AAPH)

solution was added to initiate the reaction after 30 min achieved. The excitation and

emission wavelengths were set as 485 and 520 nm, respectively. The kinetic readings

were recorded for 60 cycles with 40 seconds for each cycle. The kinetics of the

fluorescence changes were recorded immediately using software SoftMax Pro (Molecular

Devices, Sunnyvale, CA, USA). The ORAC value was calculated and expressed as

86

micromoles of Trolox equivalent per gram hydrolysates (µmol of trolox equivalent/g)

using the standard curve of Trolox.

4.3.14 Radical DPPH scavenging activity of selected hydrolysates

DPPH-free scavenging capacity of selected hydrolysates was determined

according to our previous method (Chen & Ho, 1995; Tan, Chang, & Zhang, 2017).

Twenty microliters of hydrolysates with appropriate dilutions were mixed well with 380

µL of 0.1 mM DPPH solution. The mixtures were allowed to stand at the room

temperature in the dark for 30 min. Two-hundred microliters of the mixture were added

into a 96-well plate, the absorbance was measured with a plate reader at 517 nm against

an ethanol blank. The free radical scavenging activity of the samples was expressed as

micromoles of Trolox equivalents per gram dry hydrolysates (μmol TE/g).

4.3.15 Angiotensin-converting enzyme (ACE) inhibition assay

Hydrolysates under different hydrolysis conditions were selected for ACE

inhibition assay. The ACE inhibition assay was performed according to the published

method (Wu & Ding, 2002; Zhang, Pechan, & Chang, 2018). In brief, 60 μL of ACE (10

mU/mL) dissolved in borate buffer (containing 300 mM NaCl, pH 8.3) was mixed well

with 30 μL of sample or water (for control) and stand at room temperature for 10 min.

Aliquot 60 μL of 2.5 mM hippuryl-histidyl-leucine (HHL) in the borate buffer to initiate

the reaction, and stand for 30 min at 37°C. The reaction was terminated by heating up at

85°C for 15 min. The reaction mixture was filtered through 20 um nylon syringe filter

before HPLC analysis. Hippuric acid (HA) was separated and quantified by Agilent 1260

Infinity HPLC system (Agilent Technologies, Santa Clara, CA, USA). Agilent Exlipse

87

XDB-C18 column (4.6 × 150 mm; particle size 5 µm; Agilent Technologies, Rising Sun,

MD, USA) was used and the wavelength of diode array detector (DAD) was set at

228 nm. Mobile phases solvent A (0.1% trifluoroacetic acid in water) and solvent B

(0.1% trifluoroacetic acid in acetonitrile). The elution gradient was as follow: 0–10 min,

5–60% B; 10–12 min, 60% B; 12–13 min, 60–5% B. The percentage of ACE inhibition

was calculated according to the following equation:

% inhibition = 100 – [100×(HAsample/HAcontrol)] (4.10)

where HAsample represents the content of HA when inhibitor was present;

HAcontrol represents the content of HA when inhibitor was replaced by water. The

IC50 value was defined as the concentration to inhibit 50% ACE activity and was

calculated by GraphPad Prism software based on four concentrations. Captopril was used

as positive control.

4.3.16 DPP-IV inhibition assay

This assay was carried out by using a DPP-IV inhibitor screening kit (Sigma-

Aldrich, St. Louis, MO) in accordance with the manufacture’s instruction. In brief, mix

the appropriate diluted sample solution (49 μL) with 1 μL of DPP-IV enzyme solution

and incubate at 37°C for 10 min (protect it from light). After 10 min, substrate was added

into the mixture to initiate the reaction. Measure the fluorescence (excitation wavelength

= 360, and emission wavelength = 460 nm) in a microplate reader in kinetic mode for 20

minutes at 37 °C. To correct for background in samples, include a sample blank by

omitting the DPP-IV Enzyme. Two-time points in the linear range of the plot was chose

and obtain the slope between T1 and T2. The relative inhibition rate was calculated by

the following equation:

88

Slope = (FLU2-FLU1)/(T2-T1) (4.11)

Relative inhibition rate (%) =SlopeEC−SlopeEM

SlopeEC (4.12)

where: SlopeSM = the slope of the Sample Inhibitor; SlopeEC = the slope of the Enzyme

Control. The IC50 value was defined as the concentration to inhibit 50% DPP-IV activity

and was calculated by GraphPad Prism software based on four concentrations. Sitagliptin

was used as positive control.

4.3.17 Statistical analysis

All the experiments were carried out based on a completely randomized design.

Hydrolysis processing was performed in triplicate. Data were analyzed by ANOVA using

2015 SAS (version 9.4, SAS Inc., Cary, NC, USA). Duncan’s multiple range test was

carried out to determine any significant differences (p = 0.05) between variables.

4.4 Results and discussion

4.4.1 Proximate analysis

Proximate analysis of the mixture of heads and frames was carried out. The

moisture, protein, total fat, and ash content of the mixture were 60.70%, 14.85%, 12.09%,

and 7.98%, respectively. The results indicated that the mixture of heads and frames

contains over 37% of protein (dry weight basis) which could be processed into value-

added food ingredients. Oven dried samples were used for protein, lipid and ash content

analysis, and then calculated back to wet weight basis. Because oven dried samples were

used for protein, lipid and ash content analysis, and the dried samples may have absorbed

moisture after drying. Therefore, the protein, lipid and ash content of dry weight basis did

89

not come to 100%. However, when calculated back to wet weight basis, the total number

is closer to 100%.

Table 4.1 Proximate analysis of the mixture of catfish heads and frames (w:w=3:2).

Samples Moisture (%) Protein (%) Lipid (%) Ash (%)

Mixture 60.70% ± 0.50% 14.85% ± 0.12% 12.09% ± 0.23% 7.98% ± 0.17%

Data were expressed as mean ± standard deviation (n=3). Protein, lipid and ash content

were based on wet weight basis.

4.4.2 Determination of proteolytic activity by using Azocoll and Azocasein as

substrate

Proteolytic activity of eight proteases are shown in Table 4.2. The proteolytic

activity of each protease was determined by two different methods, and the correlation

analysis of the two proteolytic activities was performed. The proteolytic activity of each

protease varied significantly, which was not surprising because that the proteases are

expected to have different proteolytic activity if the proteases are derived from different

sources. Papain, ficin and bromelain are all derived from plant sources; however, the rest

of the proteases used in this study are derived from bacteria. Thermolysin possessed the

highest proteolytic activity per g on AzocollTM and Azocasein at 70°C, followed by novo-

pro D at 60°C. The correlation between the two proteolytic activities determined by the

two different methods was 0.97, which suggesting that the two different substrate

methods did not vary significantly. However, proteases sensitivity towards these two

substrates was different, with 4 proteases having higher activities with azocasein and vice

versa. Protease activity cost of each protease is shown in Table 4.2, alcalase at 60°C

could provide 1,369,403 Azocoll unit per dollar, and followed by 1,257,669 Azocoll unit

per dollar provided by thermolysin at 70°C. The enzyme cost could provide economic

90

information for the processing industries on the basis of proteolytic activity, which is

very important as a part of the investment for the industries.

Proteases had been used based on a weight basis as reported in most of the protein

hydrolysis studies (Cai, Chang, & Lunde, 1996; Cheung & Li-Chan, 2017; Klompong,

Benjakul, Kantachote, & Shahidi, 2007; Nilsang, Lertsiri, Suphantharika, & Assavanig,

2005; Salampessy, Reddy, Phillips, & Kailasapathy, 2017). However, comparing the

protease hydrolysis on a weight basis gives little meaningful information on the

efficiency of proteases for hydrolysis. In some of the studies, which reported the protease

in enzyme unit, proteolytic activity was provided by the manufacturers (Benjakul &

Morrissey, 1997) instead of determining the proteolytic activities under the same working

conditions prior to the hydrolysis analyze experiments. In addition, literature reported

proteolytic activities were determined at different conditions and using different

substrates. Different hydrolysis conditions usually lead to different proteolytic activity,

and substrate specificity influences the protease efficiency as well. These made

comparing the results from the literature very difficult. Therefore, it is very important to

investigate and compare proteolytic activity under the same hydrolysis conditions.

Among reported publications, only two groups of researchers had reported the use of

AzocollTM to standardize proteolytic activity for salmon muscle protein hydrolysis and

comparing five proteases at the same activity level. However, our study is the first to use

both Azocoll and Azocasein as substrates to determine proteolytic activities. Even though

there are some limitations for using Azocoll and Azocasein as standard substrates, it is

convenient to get proximate activity of proteases in order to obtain references for

91

protease addition. In any rate, presenting protease activity is more accurate and repeatable

than comparing proteases on the weight basis.

92

Tab

le 4

.2

Act

ivit

y a

nd c

ost

of

eig

ht

types

of

pro

teas

es.

Enzym

es

Tem

per

ature

(°C

)

Act

ivit

y

(Azo

coll U

/g o

r

mL

)

Act

ivit

y

(Azo

case

in U

/g o

r

mL

)

Co

rrel

atio

n

coef

ficie

nt

(R2)

Enzym

e co

st

($/g

or

mL

)

Act

ivit

y c

ost

(Azo

coll U

/$)

Pap

ain

30

8048 ±

45

x

34508 ±

1618

t

0.9

6

1.3

0*

6191

40

14712 ±

120

v

60849 ±

929

r 11317

50

18060 ±

369

s 74178 ±

1315

o

13892

60

22276 ±

452

r 81092 ±

6954

n

17135

Fic

in

30

3525 ±

21

b*

94259 ±

2625

m

3.1

5*

1119

40

4109 ±

23

z 210306 ±

2496

h

1304

50

7371 ±

53

y

424508 ±

13599

f 2340

60

15062 ±

480

u

677876 ±

13082

e 4781

Bro

mela

in

30

2942 ±

34

e*

61120 ±

1204

q

3.4

5*

853

40

3274 ±

37

d*

103617 ±

7229

l 949

50

3700 ±

34

a*

221963 ±

13599

g

1072

60

3414 ±

29

c*

161467 ±

3442

i 990

Ther

mo

lysi

n

40

1.9

×10

8 ±

123

d

7.4

×10

5 ±

41313

d

652

*

291411

659509

1180982

1257669

50

4.3

×10

8 ±

225

c 1.7

×10

6 ±

58527

c

60

7.7

×10

8 ±

687

b

4.6

×10

6 ±

240996

b

70

8.2

×10

8 ±

105

a 5.6

×10

6 ±

471664

a

Neu

tras

e

30

10450 ±

17

w

1359 ±

10

b*

0.9

5/m

L

217708

40

16875 ±

178

t 2634 ±

20

a*

351562

50

27737 ±

70

p

3817 ±

117

y

577854

60

2700 ±

52

f*

1225 ±

99

c*

56250

Alc

ala

se

30

35562 ±

88

o

3369 ±

210

z

0.6

7/m

L

1045941

40

247875 ±

530

j

8711 ±

275

w

367762

50

707333 ±

367

g

33029 ±

2771

u

105572

60

917500 ±

701

f 53594 ±

8710

s 1369403

93

Tab

le 4

.2 (

conti

nued

)

Enzym

es

Tem

per

ature

(°C

)

Act

ivit

y

(Azo

coll U

/g o

r

mL

)

Act

ivit

y

(Azo

case

in U

/g o

r

mL

)

Co

rrel

atio

n

coef

ficie

nt

(R2)

Enzym

e co

st

($/g

or

mL

)

Act

ivit

y c

ost

(Azo

coll U

/$)

Pro

tam

ex

30

26150 ±

212

q

39434 ±

1099

1.5

3

17091

40

43100 ±

94

m

63626 ±

4041

p

28170

50

50583 ±

39

l 107598 ±

7516

k

33061

60

68875 ±

412

k

154828 ±

5105

j 45016

No

vo

-Pro

D

30

36975 ±

35

n

415 ±

34

e*

1.2

5

29580

40

263375 ±

164

i 1044 ±

28

d*

210700

50

664000 ±

371

h

4181 ±

59

x

531200

60

1257500 ±

353

e 10459 ±

897

v

1006000

Act

ivit

y w

as d

eter

min

ed u

nder

pH

7.2

at

dif

fere

nt

tem

per

ature

s, u

sing A

zoco

llT

M a

nd A

zoca

sein

as

stan

dar

d s

ubst

rate

s. N

eutr

ase,

Alc

als

e and N

ovo

-Pro

D a

re i

n l

iqu

id f

orm

, th

e re

st o

f pro

teas

es a

re i

n s

oli

d f

orm

. V

alu

es

mar

ked

by t

he

dif

fere

nt

lett

er w

ithin

the

sam

e co

lum

n a

re s

ignif

icantl

y d

iffe

rent

(P <

0.0

5),

and *

rep

rese

nt

the

sma

ller

valu

e dat

a w

hic

h r

ank

ed i

n t

he

seco

nd r

ound (

e.g.,

the

sam

e le

tter

wit

h a

nd w

itho

ut

* h

ad 2

6 let

ters

’ dif

fere

nce)

. * m

eans

the

pro

teas

e pri

ce w

ere

obta

ined f

rom

Sig

ma,

and t

he

oth

er

pro

teas

e pri

ce w

ere

obta

ined

fro

m N

ovo

zym

e In

c..

94

4.4.3 Degree of hydrolysis of catfish protein hydrolysates

Degree of hydrolysis is a vital index for hydrolysis processing since it influences

product yield, protein recovery rate and functional properties of the resulting

hydrolysates. In Table 4.2, the proteolytic activity based on AzocollTM was determined at

different temperatures. Protease concentrations were adjusted to 5, 10, 25, 40 and 80 AzU

per gram of protein in fish paste (protein content was determined by the Kjeldahl method)

for the hydrolysis reaction. Thermolysin had the highest activity (8.2×108 Azocoll U/g

and 5.6×106 Azocasein U/g) at 70 °C when using AzocollTM and Azocasein as the

substrates. Figure 4.1-4.8 showed the degree of hydrolysis curve for the eight proteases

(papain, ficin, bromelain, neutrase, alcalase, protamex, novo-proD and thermolysin)

applied at different proteolytic activities under different temperatures. Overall, proteases

derived from plant sources give higher degree of hydrolysis (range from 40-70%) than

that from the bacteria sources (range from 10-30%). Degree of hydrolysis exhibited a

sharp increase in the first 10 min and reached to a relative plateau-like pattern after 20

min. This observation was similar to most of the protein hydrolysis studies (Benjakul &

Morrissey, 1997; Klompong, Benjakul, Kantachote, & Shahidi, 2007; Kristinsson &

Rasco, 2000b; Rebeca, Pena-Vera, & Diaz-Castaneda, 1991; Shahidi, Han, &

Synowiecki, 1995), where the degree of hydrolysis increased rapidly in the first stage of

hydrolysis, and kept a slow increasing rate to reach a stable stage. The degree of

hydrolysis varied from 10% to 30% in most of the reported protein hydrolysis studies.

DH of 15.7% was observed when hydrolyzing the protein from defatted sunflower meal

with papain (Cai, Chang, & Lunde, 1996); DH of 32.9% was reported when hydrolyzing

95

the leather jacket insoluble protein with bromelain (Salampessy, Phillips, Seneweera, &

Kailasapathy, 2010).

The maximum DH of alcalase and novo-pro D were reported as 15-25% at

optimal pH and temperature, and 10%-20% for neutrase and protamex (Þorkelsson &

Kristinsson, 2009). In our study (Figure 4.1 to 4.8), the maximum DH of each protease

derived from bacteria was around 25%; however, proteases from plant sources produced

over 50% of DH in the current study. The highest degree of hydrolysis of hydrolysates

was observed in 120 min and at the highest enzyme concentration (80 AzU/g) for each

protease. Ficin (80 AzU/g) was the most efficient in hydrolyzing the catfish by-product,

reaching 71% DH (degree of hydrolysis) in 120 min at 30°C among all the enzymes

studied. However, the highest DH of alcalase hydrolysates was 16.48%, achieved in 120

min at 40°C. The above proteases behaved differently even at the same AzocollTM

activity, which suggesting that protease specific reaction sites on the catfish protein

chains were different from the analytical substrate. Kristinsson and Rasco (2000b)

observed similar results in their study of the hydrolysis of salmon muscle with 5

proteases (alcalase, flavourzyme, corolase PN-L, corolase 7098 and pyloric caeca

extract). This might be due to the substrate specificity differences.

96

Fig

ure

4.1

E

nzym

atic

hydro

lysi

s o

f ca

tfis

h b

y-p

roduct

(m

ixtu

re o

f fr

am

es a

nd h

eads)

wit

h p

apain

under

dif

fere

nt

tem

per

ature

and v

aryin

g a

ctiv

ity u

nit

s (A

zU/g

of

pro

tein

in s

ubst

rate

).

0102030405060

02

04

060

801

00

12

01

40

Degree of hydrolysis (%)

Tim

e (m

in)

Pap

ain

30

°C

5 A

zU10

AzU

25 A

zU40

AzU

80 A

zU

010

20

30

40

50

02

04

06

08

01

00

12

01

40

Degree of hydrolysis (%)

Tim

e (m

in)

Pap

ain

40°

C

5 A

zU10

AzU

25 A

zU40

AzU

80 A

zU

97

Fig

ure

4.1

(co

nti

nued

)

010

20

30

40

50

60

02

04

060

801

00

12

01

40

Degree of hydrolysis (%)

Tim

e (m

in)

Pap

ain

50

°C

5 A

zU10

AzU

25 A

zU40

AzU

80 A

zU

010

20

30

40

50

60

02

04

06

08

01

00

12

01

40

Degree of hydrolysis (%)

Tim

e (m

in)

Pap

ain

60

°C

5 A

zU10

AzU

25 A

zU40

AzU

80 A

zU

98

Fig

ure

4.2

E

nzym

e hydro

lysi

s o

f ca

tfis

h b

y-p

roduct

(m

ixtu

re o

f fr

am

es a

nd h

eads)

wit

h f

icin

under

dif

fere

nt

tem

per

ature

and

var

yin

g a

ctiv

ity u

nit

s (A

zU/g

of

pro

tein

in s

ubst

rate

).

020406080100

02

04

060

80

10

01

20

14

0

Degree of hydrolysis (%)

Tim

e (m

in)

Fici

n 3

0°C

5 A

zU10

AzU

25 A

zU40

AzU

80 A

zU

010

20

304050

60

70

80

02

04

06

08

01

00

12

01

40

Degree of hydrolysis (%)

Tim

e (m

in)

Fici

n 4

0°C

5 A

zU10

AzU

25 A

zU40

AzU

80 A

zU

99

Fig

ure

4.2

(co

nti

nued

)

010

20

304050

60

70

80

02

04

060

801

00

12

01

40

Degree of hydrolysis 9%)

Tim

e (m

in)

Fici

n 5

0°C

5 A

zU10

AzU

25 A

zU40

AzU

80 A

zU

010

20

304050

60

70

80

02

04

06

08

01

00

12

01

40

Degree of hydrolysis (%)

Tim

e (m

in)

Fici

n 6

0°C

5 A

zU10

AzU

25 A

zU40

AzU

80 A

zU

100

Fig

ure

4.3

E

nzym

e hydro

lysi

s o

f ca

tfis

h b

y-p

roduct

(m

ixtu

re o

f fr

am

es a

nd h

eads)

wit

h b

rom

ela

in u

nder

dif

fere

nt

tem

per

ature

and v

aryin

g a

ctiv

ity u

nit

s (A

zU/g

of

pro

tein

in s

ubst

rate

).

010

20

30

40

50

60

70

02

04

060

801

00

12

01

40

Degree of hydrolysis (%)B

rom

ela

in 3

0°C

5 A

zU10

AzU

25 A

zU40

AzU

80 A

zU

010

20

30

40

50

60

70

02

04

06

08

01

00

12

01

40

Degree of hydrolysis (%)

Tim

e (m

in)

Bro

mel

ain

40

°C

5 A

zU1

0 A

zU25

AzU

40 A

zU80

AzU

101

Fig

ure

4.3

(co

nti

nued

)

010

20

30

40

50

60

70

02

04

060

801

00

12

01

40

Degree of hydrolysis (%)

Tim

e (m

in)

Bro

me

lain

50°

C

5 A

zU10

AzU

25 A

zU40

AzU

80 A

zU

010

20

304050

60

70

80

02

04

06

08

01

00

12

01

40

Degree of hydrolysis (%)

Tim

e (m

in)

Bro

mel

ain

60

°C

5 A

zU1

0 A

zU25

AzU

40 A

zU80

AzU

102

Fig

ure

4.4

E

nzym

e hydro

lysi

s o

f ca

tfis

h b

y-p

roduct

(m

ixtu

re o

f fr

am

es a

nd h

eads)

wit

h n

eutr

ase

under

dif

fere

nt

tem

per

ature

and v

aryin

g a

ctiv

ity u

nit

s (A

zU/g

of

pro

tein

in s

ubst

rate

).

051015202530

02

04

060

801

00

12

01

40

Degree of hydrolysis 9%)

Tim

e (m

in)

Neu

tras

e 30

°C

5 A

zU10

AzU

25 A

zU40

AzU

80 A

zU

0510

15

20

25

30

35

02

04

06

08

01

00

12

01

40

Degree of hydrolysis (%)

Tim

e (m

in)

Neu

tras

e 4

0°C

5 A

zU1

0 A

zU25

AzU

40 A

zU80

AzU

103

Fig

ure

4.4

(co

nti

nued

)

0510

15

20

25

30

35

02

04

060

801

00

12

01

40

Degree of hydrolysis (%)

Tim

e (m

in)

Neu

tras

e 50

°C

5 A

zU10

AzU

25 A

zU40

AzU

80 A

zU

051015202530

02

04

06

08

01

00

12

01

40

Degree of hydrolysis (%)

Tim

e (m

in)

Neu

tras

e 6

0°C

5 A

zU1

0 A

zU25

AzU

40 A

zU80

AzU

104

Fig

ure

4.5

E

nzym

e hydro

lysi

s o

f ca

tfis

h b

y-p

roduct

(m

ixtu

re o

f fr

am

es a

nd h

eads)

wit

h a

lcala

se u

nd

er d

iffe

rent

tem

per

ature

and

var

yin

g a

ctiv

ity u

nit

s (A

zU/g

of

pro

tein

in s

ubst

rate

).

0510

152025

30

35

40

02

04

060

801

00

12

01

40

Degree of hydrolysis (%)

Tim

e (m

in)

Alc

alas

e 30

°C

5 A

zU10

AzU

25 A

zU40

AzU

80 A

zU

051015

20

02

04

06

08

01

00

12

01

40

Degree of hydrolysis (%)

Tim

e (m

in)

Alc

alas

e 4

0°C

5 A

zU10

AzU

25 A

zU40

AzU

80 A

zU

105

Fig

ure

4.5

(co

nti

nued

)

051015202530

02

04

060

801

00

12

01

40

Degree of hydrolysis (%)

Tim

e (m

in)

Alc

alas

e 50

°C

5 A

zU10

AzU

25 A

zU40

AzU

80 A

zU

051015202530

02

04

06

08

01

00

12

01

40

Degree of hydrolysis (%)

Tim

e (m

in)

Alc

alas

e 6

0°C

5 A

zU10

AzU

25 A

zU40

AzU

80 A

zU

106

Fig

ure

4.6

E

nzym

e hydro

lysi

s o

f ca

tfis

h b

y-p

roduct

(m

ixtu

re o

f fr

am

es a

nd h

eads)

wit

h p

rota

mex u

nder

dif

fere

nt

tem

per

ature

and v

aryin

g a

ctiv

ity u

nit

s (A

zU/g

of

pro

tein

in s

ubst

rate

).

051015202530

02

04

060

801

00

12

01

40

Degree of hydrolysis (%)

Tim

e (m

in)

Pro

taM

ex 3

0°C

5 A

zU10

AzU

25 A

zU40

AzU

80 A

zU

051015202530

02

04

06

08

01

00

12

01

40

Degree of hydrolysis (%)

Tim

e (m

in)

Pro

taM

ex 4

0°C

5 A

zU1

0 A

zU25

AzU

40 A

zU80

AzU

107

Fig

ure

4.6

(co

nti

nued

)

0510

15

20

25

30

35

02

04

060

801

00

12

01

40

Degree of hydrolysis (%)

Tim

e (m

in)

Pro

taM

ex 5

0°C

5 A

zU10

AzU

25 A

zU40

AzU

80 A

zU

0510

15

20

25

02

04

06

08

01

00

12

01

40

Degree of hydrolysis (%)

Tim

e (m

in)

Pro

taM

ex 6

0°C

5 A

zU1

0 A

zU25

AzU

40 A

zU80

AzU

108

Fig

ure

4.7

E

nzym

e hydro

lysi

s o

f ca

tfis

h b

y-p

roduct

(m

ixtu

re o

f fr

am

es a

nd h

eads)

wit

h N

ovo

-Pro

D u

nder

dif

fere

nt

tem

per

ature

and v

aryin

g a

ctiv

ity u

nit

s (A

zU/g

of

pro

tein

in s

ubst

rate

).

0510

15

20

25

30

35

02

04

060

801

00

12

01

40

Degree of hydrolysis (%)

Tim

e (m

in)

No

vo-P

roD

30

°C

5 A

zU10

AzU

25 A

zU40

AzU

80 A

zU

0510

15

20

25

02

04

06

08

01

00

12

01

40

Degree of hydrolysis (%)

Tim

e (m

in)

No

vo-P

roD

40

°C

5 A

zU10

AzU

25 A

zU40

AzU

80 A

zU

109

Fig

ure

4.7

(co

nti

nued

)

0510

15

20

25

02

04

060

801

00

12

01

40

Degree of hydrolysis (%)

Tim

e (m

in)

No

vo-P

roD

50

°C

5 A

zU10

AzU

25 A

zU40

AzU

80 A

zU

0510

15

20

25

02

04

06

08

01

00

12

01

40

Degree of hydrolysis (%)

Tim

e (m

in)

No

vo-P

roD

60

°C

5 A

zU10

AzU

25 A

zU40

AzU

80 A

zU

110

Fig

ure

4.8

E

nzym

e hydro

lysi

s o

f ca

tfis

h b

y-p

roduct

(m

ixtu

re o

f fr

am

es a

nd h

eads)

wit

h t

her

mo

lysi

n u

nder

dif

fere

nt

tem

per

ature

and v

aryin

g a

ctiv

ity u

nit

s (A

zU/g

of

pro

tein

in s

ubst

rate

).

051015202530

02

04

060

801

00

12

01

40

Degree of hydrolysis (%)

Tim

e (m

in)

Ther

mo

lysi

n 4

0°C

5 A

zU10

AzU

25 A

zU40

AzU

80 A

zU

0510

15

20

25

30

35

02

04

06

08

01

00

12

01

40

Degree of hydrolysis (%)

Tim

e (m

in)

Ther

mo

lysi

n 5

0°C

5 A

zU10

AzU

25 A

zU40

AzU

80 A

zU

111

Fig

ure

4.8

(co

nti

nued

)

0510

15

20

25

30

35

02

04

060

801

00

12

01

40

Degree of hydrolysis (%)

Tim

e (m

in)

Ther

mo

lysi

n 6

0°C

5 A

zU10

AzU

25 A

zU40

AzU

80 A

zU

0510

152025

30

35

40

02

04

06

08

01

00

12

01

40

Degree of hydrolysis (%)

Tim

e (m

in)

Ther

mo

lysi

n 7

0°C

5 A

zU10

AzU

25 A

zU40

AzU

80 A

zU

112

4.4.4 Kinetic analysis of hydrolysis processing

Kinetic analysis for the hydrolysis processing was carried out using the Peleg’s model at

various temperatures. Peleg constants K1 and K2 are shown in Table 4.3, K1 relates to

degree of hydrolysis at very beginning, and Peleg capacity constant K2 relates to

maximum of degree of hydrolysis. Degree of hydrolysis exhibited an overall increase as

time was extended. The hydrolysis curves indicated that ficin gave the highest degree of

hydrolysis (71%) at 30°C. In the literature, single temperature or protease concentration

had been used for hydrolysis processing for most of the protein hydrolysis studies (Cai,

Chang, & Lunde, 1996; Hathwar, Bijinu, Rai, & Narayan, 2011; Hoyle & Merritt, 1994;

Liaset, Lied, & Espe, 2000), thus making the kinetic analysis impossible. For the study

that had conducted the kinetic analysis on hydrolysis of salmon muscle with 5 proteases

as was discussed above (Kristinsson & Rasco, 2000b), the model applied was different

from our current study, which makes the comparison difficult. Though the model applied

was not the same, the changing patterns of degree of hydrolysis were similar. As shown

in Table 4.3, the lowest K1 and K2 were found in ficin hydrolysis (50 and 40°C,

respectively). The hydrolysis curves fit the modified Peleg model well, having R2 all

higher than 0.91. Kinetic analysis had been performed on the hydrolysis of soy protein,

and Michaelis-Menten model was used (Jens Adler-Nissen, 1977), however, degree of

hydrolysis could not be predicted through this protease reaction model for the soy protein

study.

113

Table 4.3 Hydrolysis kinetics of 8 enzymes at different temperatures and enzyme

concentrations.

Temperature (°C) Hydrolysis time (min) K1 K2 R2

Papain 30 10 0.38 0.80 0.93

20 0.58 0.73 0.95

40 0.95 1.69 0.94

60 1.41 1.98 0.99

120 2.77 3.76 0.99

40 10 0.37 0.93 0.99

20 0.66 1.58 0.97

40 1.21 3.53 0.99

60 1.72 4.51 0.98

120 3.35 8.99 0.99

50 10 0.33 0.79 0.93

20 0.55 1.01 0.96

40 0.99 2.27 0.99

60 1.46 3.24 0.98

120 2.94 4.77 0.97

60 10 0.27 0.78 0.96

20 0.52 1.25 0.97

40 0.98 2.58 0.96

60 1.40 3.77 0.94

120 2.74 7.07 0.94

Ficin 30 10 0.19 0.09 0.91

20 0.33 0.15 0.95

40 0.62 0.33 0.91

60 0.93 0.35 0.91

120 1.80 0.44 0.93

40 10 0.19 0.08 0.97

20 0.35 0.23 0.91

40 0.67 0.48 0.92

60 0.99 0.44 0.93

120 1.95 0.80 0.97

50 10 0.18 0.10 0.94

20 0.36 0.18 0.98

40 0.69 0.37 0.98

60 1.04 0.42 0.95

120 2.03 0.72 0.92

114

Table 4.3 (continued)

Temperature (°C) Hydrolysis time (min) K1 K2 R2

Ficin 60 10 0.21 0.09 0.97

20 0.36 0.09 0.98

40 0.69 0.11 0.95

60 1.01 0.12 0.94

120 1.97 0.30 0.96

Bromelain 30 10 0.23 0.11 0.97

20 0.43 0.23 0.94

40 0.83 0.34 0.97

60 1.22 0.30 0.99

120 2.38 0.79 0.91

40 10 0.22 0.34 0.99

20 0.43 0.59 0.99

40 0.83 1.18 0.99

60 1.21 1.49 0.99

120 2.38 2.82 0.99

50 10 0.22 0.18 0.98

20 0.41 0.46 0.95

40 0.79 0.79 0.95

60 1.13 1.15 0.93

120 2.23 2.19 0.94

60 10 0.20 0.15 0.98

20 0.39 0.33 0.94

40 0.76 0.57 0.98

60 1.11 0.77 0.96

120 2.18 1.42 0.97

Neutrase 30 10 0.81 12.40 0.97

20 1.63 14.03 0.92

40 2.90 28.75 0.94

60 3.70 45.12 0.96

120 7.69 65.62 0.93

40 10 0.79 2.76 0.92

20 1.25 4.92 0.99

40 2.02 8.77 0.99

60 2.67 12.58 0.99

120 5.20 21.13 0.98

115

Table 4.3 (continued)

Temperature (°C) Hydrolysis time (min) K1 K2 R2

50 10 0.62 4.93 0.96

20 1.16 2.93 0.99

40 2.08 2.81 0.94

60 2.86 4.83 0.94

120 5.55 7.75 0.95

Neutrase 60 10 0.87 15.78 0.99

20 1.37 20.88 0.94

40 2.40 42.76 0.99

60 2.97 42.76 0.99

120 5.52 78.51 0.99

Alcalase 30 10 0.37 10.09 0.95

20 0.78 9.34 0.94

40 1.54 13.50 0.93

60 2.19 18.17 0.91

120 4.20 29.55 0.93

40 10 0.97 11.45 0.97

20 1.82 13.38 0.95

40 3.43 19.82 0.91

60 4.95 21.96 0.91

120 9.55 40.45 0.91

50 10 0.72 5.58 0.92

20 1.28 10.78 0.91

40 2.36 17.69 0.91

60 1.28 10.78 0.91

120 6.42 42.37 0.93

60 10 0.64 6.59 0.98

20 1.22 9.61 0.96

40 2.21 17.87 0.98

60 3.03 25.58 0.98

120 5.63 46.05 0.98

Protamex 30 10 0.96 7.30 0.92

20 1.26 14.93 0.99

40 2.30 29.63 0.99

60 3.19 40.66 0.99

120 5.70 76.54 0.99

116

Table 4.3 (continued)

Temperature

(°C)

Hydrolysis time (min) K1 K2 R2

40 10 0.45 20.67 0.99

20 0.52 39.63 0.97

40 1.29 61.10 0.99

60 2.60 59.48 0.99

120 4.99 101.58 0.99

50 10 0.50 8.86 0.98

20 0.91 15.22 0.98

40 1.88 23.43 0.99

60 2.70 28.39 0.98

120 5.23 50.40 0.98

Protamex 60 10 1.01 4.27 0.92

20 1.79 6.95 0.91

40 3.20 13.30 0.92

60 4.25 16.06 0.94

120 8.08 28.46 0.93

Novo-pro D 30 10 0.42 11.80 0.93

20 0.99 11.99 0.97

40 1.77 20.11 0.984

60 2.37 28.53 0.98

120 4.60 49.83 0.96

40 10 1.70 8.73 0.92

20 2.08 17.39 0.95

40 3.83 26.73 0.96

60 5.23 33.18 0.91

120 9.63 56.60 0.92

50 10 0.97 10.56 0.92

20 1.82 17.02 0.93

40 3.49 30.27 0.92

60 4.89 39.03 0.92

120 8.78 68.84 0.94

60 10 1.19 8.72 0.92

20 2.13 16.30 0.91

40 3.87 27.64 0.93

60 5.74 33.91 0.93

120 8.91 66.45 0.93

117

Table 4.3 (continued)

Temperature (°C) Hydrolysis time (min) K1 K2 R2

Thermolysin 40 10 1.03 6.07 0.91

20 1.58 11.64 0.96

40 2.91 18.33 0.95

60 3.81 20.31 0.98

120 6.73 24.96 0.98

50 10 0.36 7.16 0.97

20 0.68 12.81 0.94

40 1.61 14.36 0.96

60 2.16 21.99 0.95

120 4.55 28.04 0.99

60 10 0.41 4.90 0.99

20 0.74 8.18 0.98

40 1.54 11.08 0.97

60 2.10 16.78 0.99

120 4.19 28.26 0.99

70 10 1.15 4.65 0.99

20 1.31 8.88 0.99

40 1.93 15.23 0.95

60 2.52 20.28 0.99

120 4.63 36.10 0.98

Price were calculated according to the linear model obtained from kinetic analysis.

4.4.5 Predicted cost of protease for approaching certain DH

Developing a model to estimate investment is crucial for the food industries to

make any production decision. One study had reported (Kristinson & Rasco, 2000b) the

estimated enzyme cost, however, the cost was calculated based on enzyme price back to

almost twenty years ago. Nearly all reported studies on protein hydrolysis claimed that

they had developed a potential method to utilize by-products for industries (Benjakul &

118

Morrissey, 1997; Cheung & Chan, 2017; Diniz & Martin, 1997; Guerard, Dufosse, De La

Broise, & Binet, 2001; Ojha, Alvarez, Kumar, O'Donnell, & Tiwari, 2016; Zhou, et al.,

2016); however, no estimate cost had been provided.

In our current study, protease concentrations for approaching a certain degree of

hydrolysis level was calculated from the kinetic model, and the results are shown in

Table 4.4. The price for food grade proteases (neutrase, alcalase, protamex, novo-ProD)

was obtained from Novozyme Inc. (Franklinton, NC, USA). The price for other proteases

were obtained from Sigma (St. Louis, MO, USA). This calculation could be applied for

any desired degree of hydrolysis. However, extrapolating out of the range of the

experimental conditions might result in inaccurate cost. Nonetheless, this calculation

could be performed for any reaction time and temperature within the range of experiment

points. The protease cost is presented in Table 4.3, for reaching 15 and 20% of DH after

hydrolysis for 120 min at different temperatures were calculated based on the linear

relationship derived from the kinetic model. The lowest protease activity needed for

hydrolyzing 100 g of proteins in the substrate and approaching DH to 15% and 20% was

found in ficin hydrolysis for 120 min at 60°C (3.0 and 5.0 AzU/g, respectively).

However, the lowest protease cost to reach DH of 15% and 20% were found in

thermolysin hydrolysis for 120 min at 50°C (0.0003 and 0.0006 $/100 g of protein,

respectively).

4.4.6 Sodium-dodecyl-sulfate gel electrophoresis (SDS-PAGE)

Figure 4.9 to Figure 4.15 showed the selected SDS-PAGE patterns of

hydrolysates. All the hydrolysates hydrolyzed by papain, ficin and bromelain (plant

sources derived) showed a similar pattern (only SDS-PAGE of papain hydrolysates are

119

shown). No myosin heavy chain (around 205 KDa) and light chain (about 23 KDa) were

observed, and actin at about 43 KDa disappeared as well. All major bands exhibited in

the control (no protease treatment) group were all degraded, and this observation was

consistent with the degree of hydrolysis results. However, all hydrolysates obtained from

the reactions of alcalase, neutrase, protamex, novo-proD and thermolysin maintained the

integrity of the myofibrillar proteins at different degrees. Myosin heavy chain was

degraded into small peptides for all hydrolysates hydrolyzed by neutrase (80AzU/g) at

30, 40, 50 and 60°C (gels for 40, 50 and 60°C are not shown). Moreover, a new band was

formed at 5 KDa for the hydrolysates hydrolyzed by neutrase at 30 and 40°C.

In the alcalase hydrolysates, actin broken into small proteins first, and followed

by myosin heavy chain. And this fragmentation phenomenon was supported by both

SDS-PAGE gels and DH results, because the actin band disappeared first in the

hydrolysates, which possessed a lower DH (hydrolysates hydrolyzed at 40°C). Unlike

alcalase hydrolysis, thermolysin hydrolysis showed an opposite digestion order, and

thermolysin cleaves at the N-terminus of Leu, Phe, Val, Ile, Ala and Met. Myosin broke

into small proteins first and followed by actin. This point could be supported by the SDS-

PAGE gel of thermolysin hydrolysates at 50°C, where the myosin heavy chain degraded

first but actin still maintained its integrity. Sen and coworkers (Sen, Sripathy, Lahiry,

Sreenivasan, & Subrahmanyan, 1962) optimized the hydrolysis parameters based on

solubility and peptide chain length; and they found that the hydrolysates from an

optimized process had a low level of free tryptophan (Sripathy, Sen, Lahiry, Sreenivasan,

& Subrahmanyan, 1962). After adjusting the hydrolysis pH and extending the hydrolysis

time, new hydrolysates with higher content of tryptophan were obtained (Sripathy,

120

Kadkol, Sen, Swaminathan, & Lahiry, 1963). This suggesting that the individual amino

acids were not released equally fast during the hydrolysis process. All proteases used in

this study are endoproteases, which break the peptide bonds on nonterminal amino acids.

In other words, endoproteases can only cut the amino acids within the molecules, thus

endoproteases cannot break the proteins into monomers. Limited studies (and none for

catfish by-products) had reported the degree of hydrolysis and SDS-PAGE gel pattern at

the same time (Cai, Chang, & Lunde, 1996; Liceaga-Gesualdo & Li-Chan, 1999; Ndez-

Diaaz & Fera, 1999; Theodore & Kristinsson, 2007). However, SDS-PAGE patterns of

hydrolysates helped us understand how the protein composition changed during

hydrolysis of proteins derived from catfish by-products.

121

Tab

le 4

.4

Calc

ula

ted e

nzym

e ac

tivit

y a

nd c

ost

to

rea

ch 1

5 a

nd 2

0%

DH

when h

ydro

lyzin

g c

atfi

sh b

y-p

roduct

at

pH

7.2

.

Pro

teas

es

Tem

per

ature

(°C

)

Hydro

lysi

s ti

me

(min

)

Pro

teas

e ac

tivit

y n

eed

ed

(AzU

/ 100 g

of

pro

tein

in s

ubst

rate

)

Pro

teas

e co

st

($/1

00 g

of

pro

tein

in s

ubst

rate

)

15%

20%

15%

20%

Pap

ain

30

120

41

72

0.0

066

0.0

116

40

120

104

193

0.0

092

0.0

171

50

120

53

94

0.0

046

0.0

081

60

120

76

134

0.0

023

0.0

041

Fic

in

30

120

4.3

7.2

0.0

038

0.0

064

40

120

7.9

13

0.0

007

0.0

011

50

120

7.3

12

0.0

006

0.0

010

60

120

3.0

5.0

0.0

001

0.0

002

122

Tab

le 4

.4 (

conti

nued

)

Pro

teas

es

Tem

per

ature

(°C

)

Hydro

lysi

s ti

me

(min

)

Pro

teas

e ac

tivit

y n

eed

ed

(AzU

/ 100 g

of

pro

tein

in s

ubst

rate

)

Pro

teas

e co

st

($/1

00 g

of

pro

tein

in s

ubst

rate

)

15%

20%

15%

20%

Bro

mela

in

30

120

8.2

14

0.0

096

0.0

164

40

120

30

50

0.0

307

0.0

511

50

120

22

38

0.0

205

0.0

354

60

120

14

24

0.0

141

0.0

243

Neu

tras

e 30

120

1522

NA

0.0

520

NA

40

120

311

756

0.0

175

0.0

425

50

120

120

316

0.0

109

0.0

287

60

120

1211

3163

0.4

252

1.1

106

123

Tab

le 4

.4 (

conti

nued

)

Pro

teas

es

Tem

per

ature

(°C

)

Hydro

lysi

s ti

me

(min

)

Pro

teas

e ac

tivit

y n

eed

ed

(AzU

/ 100 g

of

pro

tein

in s

ubst

rate

)

Pro

teas

e co

st

($/1

00 g

of

pro

tein

in s

ubst

rate

)

15%

20%

15%

20%

Alc

ala

se

30

120

379

777

0.0

072

0.0

147

40

120

1654

NA

0.0

045

NA

50

120

760

2689

0.0

007

0.0

026

60

120

723

1943

0.0

005

0.0

014

Pro

tam

ex

30

120

1214

3322

0.0

710

0.1

944

40

120

1450

3376

0.0

418

0.0

973

50

120

745

1822

0.0

225

0.0

551

60

120

726

NA

0.0

161

NA

124

Tab

le 4

.4 (

conti

nued

)

Pro

teas

es

Tem

per

ature

(°C

)

Hydro

lysi

s ti

me

(min

)

Pro

teas

e ac

tivit

y n

eed

ed

(AzU

/ 100 g

of

pro

tein

in s

ubst

rate

)

Pro

teas

e co

st

($/1

00 g

of

pro

tein

in s

ubst

rate

)

15%

20%

15%

20%

No

vo

-pro

D

30

120

673

1463

0.0

228

0.0

495

40

120

2388

NA

0.0

113

NA

50

120

2138

NA

0.0

040

NA

60

120

2151

NA

0.0

021

NA

Ther

mo

lysi

n

40

120

473

1962

0.0

006

0.0

024

50

120

377

814

0.0

003

0.0

007

60

120

362

741

0.0

003

0.0

006

70

120

490

1072

0.0

016

0.0

036

Pri

ce w

ere

calc

ula

ted a

cco

rdin

g t

o t

he

linear

mo

del

obta

ined

fro

m k

inet

ic a

naly

sis.

125

Fig

ure

4.9

C

atfi

sh p

rote

in a

nd p

epti

de

pat

tern

changes

upo

n P

rota

mex h

ydro

lysi

s at

50°C

.

MH

C:

myo

sin h

eavy c

hain

; A

CT

: ac

tin.

126

Fig

ure

4.1

0

Cat

fish

pro

tein

and p

epti

de

pat

tern

changes

upo

n N

ovo

-pro

D h

ydro

lysi

s at

60°C

.

MH

C:

myo

sin h

eavy c

hain

; A

CT

: ac

tin.

127

Fig

ure

4.1

1

Cat

fish

pro

tein

and p

epti

de

pat

tern

changes

upo

n P

apain

hydro

lysi

s at

40°C

.

MH

C:

myo

sin h

eavy c

hain

; A

CT

: ac

tin.

128

Fig

ure

4.1

2

Cat

fish

pro

tein

and p

epti

de

pat

tern

changes

upo

n A

cala

se h

ydro

lysi

s at

30°C

.

MH

C:

myo

sin h

eavy c

hain

; A

CT

: ac

tin.

129

Fig

ure

4.1

3

Cat

fish

pro

tein

and p

epti

de

pat

tern

changes

upo

n N

eutr

ase

hydro

lysi

s at

30

°C.

MH

C:

myo

sin h

eavy c

hain

; A

CT

: ac

tin.

130

Fig

ure

4.1

4

Cat

fish

pro

tein

and p

epti

de

pat

tern

changes

upo

n T

her

mo

lysi

n h

ydro

lysi

s at

50°C

.

MH

C:

myo

sin h

eavy c

hain

; A

CT

: ac

tin.

131

Fig

ure

4.1

5

Cat

fish

pro

tein

and p

epti

de

pat

tern

changes

upo

n f

icin

hydro

lysi

s at

60°C

.

MH

C:

myo

sin h

eavy c

hain

; A

CT

: ac

tin.

132

4.4.7 Emulsion capacity and stability

Emulsion activity index (EAI) and emulsion stability index (ESI) of selected

hydrolysates were determined, and soy protein isolate (SPI) was used as a positive

control. Selected results are shown in Figure 4.16 A-D. In general, hydrolysates with

lower degrees of hydrolysis (DH) exhibited higher emulsion capacities. This conclusion

was in accordance with the previous report that hydrolysates with limited degrees of

hydrolysis possessed higher emulsion capacities (Hall, Jones, O'Haire, & Liceaga, 2017).

Higher amount of hydrophobic peptides and larger molecular mass peptides would result

in a more stable emulsion system. Hydrolysates with high degrees of hydrolysis often

indicated lower emulsion capacity and stability (Ghribi, et al., 2015). Adsorption of

peptides on the surface of oil droplets during homogenization and the formation of a

membrane that blocks coalescence of the oil droplet could produce a stable emulsion

system (Mutilangi, Panyam, & Kilara, 1996). In general, it can be expected that a higher

degree of hydrolysis would lead to a lower emulsion capacity and stability (Ghribi, et al.,

2015). Adsorption of peptide chains on the surface of oil droplets during homogenization

and the formation of a membrane that blocks coalescence of the oil droplet could result a

stable emulsion system. Hydrolysates of peptides are surface active material, which could

help oil-in-water emulsion system formation since they possessed both hydrophobic and

hydrophilic groups. Peptides patterns of very small size could migrate quickly at the

interface which would make the emulsion system unstable. In addition, amphiphilicity of

peptides is the other factor which has the significant effect on the emulsifying capacity.

The amino acid sequence in an oil-in-water emulsion system had been analyzed, and it

was discovered that the amphiphilicity of peptides is more important than that of peptides

133

molecular size to emulsion capacity (Rahali, Chobert, Haertle, & Gueguen, 2000).

In this study, the lowest EAI and ESI were found at pH 4, which was in

accordance with the previous report that the lowest EAI and ESI of yellow stripe trevally

protein were observed at pH 4 (Klompong, Benjakul, Kantachote, & Shahidi, 2007).

Most of the catfish protein hydrolysates exhibited lower EAI and ESI than that of soy

protein isolate (SPI). However, only three hydrolysates exhibited comparable ESI and

EAI with SPI, and the three hydrolysates are hydrolyzed by Novo-Pro D (5 and 25

AzU/g, 10 and 20 min, respectively) and thermolysin (25 AzU/g, 20min) at 30 and 60 °C,

respectively. Emulsion property of hydrolysates from blue whiting was studied, and

exhibited much lower EAI and ESI than the current study (Egerton, Culloty, Whooley,

Stanton, & Ross, 2018). As noted in this study, protamex hydrolysates showed similar

DH and SDS-PAGE patterns with novo-ProD, but the emulsion capacity of prota-max

(data are not shown) was significantly higher than that of novo-ProD. Emulsifying

properties and stability of hydrolysates varied significantly even though the degrees of

hydrolysis were similar. This indicated that EAI and ESI are not only rely on the

molecular size of peptides but also related to the composition and amphiphilicity

structure of peptides.

134

Fig

ure

4.1

6

Em

uls

ifyin

g a

ctiv

ity i

ndex (

EA

I) a

nd e

mu

lsio

n s

tabil

ity i

ndex (

ES

I) o

f ca

tfis

h b

y-p

roduct

s hydro

lysa

tes

pre

par

ed

usi

ng N

ovo

-Pro

D (

A a

nd B

) an

d t

her

mo

lysi

n (

C a

nd D

) at

30 a

nd 6

0°C

, re

spec

tively

.

0.0

0

50.0

0

100

.00

150

.00

200

.00

250

.00

02

46

81

01

2

EAI (m2/g)

pH

SP

I5

AzU

/g 1

0m

in2

5 A

zU/g

20 m

in

40 A

zU/g

40m

in8

0 A

zU/g

120

min

05

10

15

20

25

30

35

40

45

02

46

81

01

2

ESI (%)

pH

SP

I5

AzU

/g 1

0m

in2

5 A

zU/g

20 m

in

40 A

zU/g

40m

in8

0 A

zU/g

120

min

A

B

135

Fig

ure

4.1

6 (

conti

nued

)

0.0

0

50.0

0

100

.00

150

.00

200

.00

250

.00

02

46

81

01

2

EAI (m2/g)

pH

SP

I5

AzU

/g 1

0m

in2

5 A

zU/g

20 m

in

40 A

zU/g

40m

in8

0 A

zU/g

120

min

C

05

10

15

20

25

30

35

40

45

02

46

81

01

2

ESI (%)

pH

SP

I5

AzU

/g 1

0m

in2

5 A

zU/g

20 m

in

40 A

zU/g

40m

in8

0 A

zU/g

120

min

D

136

4.4.8 Foaming capacity and stability

Hydrolysates with good foaming capacity and stability were selected and shown

in Figure 4.17 (A-H). In general, the foaming capacity of the selected hydrolysates were

higher than that of SPI, however, most of the hydrolysates exhibited much lower foaming

stability than that of SPI. This might be because that small peptides do not have enough

strength to maintain a stable foam (Shahidi, Han, & Synowiecki, 1995). Foaming

property of the silver carp protamex hydrolysates exhibited higher foaming capacity and

stability than other hydrolysates (Liu et al., 2014). High molecular weight of protein

usually correlated with potent foaming capacity and stability.

To develop a good foam, proteins or peptides must have to migrate quickly to the

air-water interface. Flexible protein domains would enhance the foaming stability. In the

current study, foaming capacity and stability of hydrolysates was affected by pH

significantly (p<0.05). The lowest foaming capacity and stability were observed at pH 4.

This was in agreement with the previous report on yellow strip trevally (Klompong,

Benjakul, Kantachote, & Shahidi, 2007). Good foaming property and stability were found

in alcalase hydrolysate (5 AzU/g, 10min at 30°C), Novo-Pro D hydrolysates (5 AzU/g,

10 min at 40, 50 and 60°C), papain hydrolysates (5 AzU/g, 10min at 30°C), protamex

hydrolysates (5 AzU/g, 10min at 30, 40, 50 and 60°C), and thermolysin hydrolysates (5

AzU/g, 10min at 30, 40, 50 and 60°C). All of those hydrolysates had low degrees of

hydrolysis (< 15%) except papain hydrolysate, and the SDS-PAGE patterns of the papain

hydrolysates had more smaller peptides than that of others. Good foaming capacity and

stability not only related to the degree of hydrolysis but also related to ionic repulsion

(Klompong, Benjakul, Kantachote, & Shahidi, 2007). Good foaming capacity had been

137

observed in capelin protein hydrolysate (Shahidi, Han, & Synowiecki, 1995) and β-

lactoglobulin (Corzo-Martínez, Moreno, Villamiel, Patino, & Sánchez, 2017) with

limited degrees of hydrolysis.

4.4.9 Antioxidant activity of selected hydrolysates

Antioxidant activity usually evaluated by more than one method (Tan, Chang, &

Zhang, 2017; Zhang, Pechan, & Chang, 2018) since antioxidant activity assay are based

on different reaction mechanisms (Prior, Wu, & Schaich, 2005). Numerous methods have

been used to determine antioxidant activity in literature. Among which, DPPH assay and

oxygen radical absorbance capacity (ORAC) are the two methods most commonly used.

DPPH assay is based on the electron transfer mechanism, and ORAC assay is depends on

the hydrogen transfer mechanism. DPPH assay was used to determine the antioxidant

activity of cod hydrolysates (Farvin, Andersen, Otte, Nielsen, Jessen, & Jacobsen, 2016).

ORAC assay was applied to study the antioxidant activity of cod protein hydrolysates

(Halldorsdottir, Sveinsdottir, Gudmundsdottir, Thorkelsson, & Kristinsson, 2014).

The ORAC values and results from DPPH assay are shown in Table 4.5 and Table

4.6, respectively. ORAC values of selected hydrolysates ranging from 157.9 to 2204.1

µmol TE/g, and DPPH values of selected hydrolysates varied from 35.4 to 204.7 µmol

TE/g. Antioxidant activity of cod bone mince hydrolysates have been reported ranging

from 435 to 528 µmol TE/g, which was similar with our current study (Halldorsdottir,

Sveinsdottir, Gudmundsdottir, Thorkelsson, & Kristinsson, 2014). In general,

hydrolysates with higher degree of hydrolysis (DH) exhibited stronger antioxidant

activity in both assay, and hydrolysates with stronger antioxidant activity exhibited

stronger ACE inhibition capacity. Pearson correlation between ACE inhibition and DPPH

138

values was analyzed, a negative correlation was observed with r equal to 0.48 at 0.05

significance level. And no significance correlation between ACE inhibition and ORAC

value was observed. Antioxidant activity was observed in selected hydrolysates but

showed no inhibition capacity against ACE, which indicated that antioxidant activity is

not the only factor for potent ACE inhibition capacity.

139

Fig

ure

4.1

7

Fo

am

ing p

roper

ty a

nd s

tabil

ity o

f ca

tfis

h b

y-p

roduct

s hydro

lysa

tes.

Alc

ala

se (

A a

nd B

) at

30°C

. B

ars

repre

sent

stan

dar

d d

evia

tio

ns

fro

m t

rip

licat

e det

erm

inat

ions.

So

y p

rote

in i

sola

te (

SP

I) w

as u

sed a

s

po

siti

ve

contr

ol.

0.00

%

50.0

0%

100.

00%

150.

00%

200.

00%

02

46

810

12

Foaming capacity (%)

pH

5 A

zU/g

10m

in25

AzU

/g 2

0 m

in40

AzU

/g 4

0min

80 A

zU/g

12

0 m

inSP

I

A

0.00

%

20.0

0%

40.0

0%

60.0

0%

80.0

0%

100.

00%

120.

00%

02

46

810

12

Foaming stability (%)

pH

5 A

zU/g

10m

in25

AzU

/g 2

0 m

in40

AzU

/g 4

0min

80 A

zU/g

120

min

SPI

BA

B

140

Fig

ure

4.1

7 (

conti

nued

)

No

vo

-pro

D (

C a

nd D

) at

30°C

. B

ars

repre

sent

stan

dar

d d

evia

tio

ns

fro

m t

rip

licat

e det

erm

inat

ions.

So

y p

rote

in i

sola

te (

SP

I) w

as

use

d a

s po

siti

ve

contr

ol.

0.00

%

50.0

0%

100.

00%

150.

00%

200.

00%

02

46

810

12

Foaming capacity (%)

pH

5 A

zU/g

10

min

25 A

zU/g

20

min

40 A

zU/g

40m

in

80

AzU

/g 1

20 m

inSP

I

0.00

%

20.0

0%

40.0

0%

60.0

0%

80.0

0%

100.

00%

120.

00%

02

46

81

01

2

Foaming stability (%)

pH

5 A

zU/g

10m

in25

AzU

/g 2

0 m

in40

AzU

/g 4

0min

80 A

zU/g

120

min

SPI

D

C

C

D

141

Fig

ure

4.1

7 (

conti

nued

)

Pap

ain

(E

and F

) at

30°C

. B

ars

repre

sent

stan

dar

d d

evia

tio

ns

fro

m t

rip

lica

te d

eter

min

atio

ns.

So

y p

rote

in i

sola

te (

SP

I) w

as u

sed a

s

po

siti

ve

contr

ol.

0.00

%

50.0

0%

100.

00%

150.

00%

200.

00%

02

46

81

01

2

Foaming capacity (%)

pH

5 A

zU/g

10m

in25

AzU

/g 2

0 m

in40

AzU

/g 4

0min

80 A

zU/g

120

min

SPI

0.00

%

20.0

0%

40.0

0%

60.0

0%

80.0

0%

100.

00%

120.

00%

02

46

81

01

2

Foaming stability (%)

pH

5 A

zU/g

10m

in25

AzU

/g 2

0 m

in40

AzU

/g 4

0min

80 A

zU/g

120

min

SPI

E

F

E

F

142

Fig

ure

4.1

7 (

conti

nued

)

Ther

mo

lysi

n (

G a

nd H

) at

50°C

. B

ars

repre

sent

stan

dar

d d

evia

tio

ns

fro

m t

rip

licat

e det

erm

inat

ions.

Soy p

rote

in i

sola

te (

SP

I) w

as

use

d a

s po

siti

ve

contr

ol.

0.00

%

50.0

0%

100.

00%

150.

00%

200.

00%

02

46

81

01

2

Foaming capacity (%)

pH

5 A

zU/g

10m

in25

AzU

/g 2

0 m

in40

AzU

/g 4

0min

80 A

zU/g

120

min

SPI

0.00

%

20.0

0%

40.0

0%

60.0

0%

80.0

0%

100.

00%

120.

00%

02

46

81

01

2

Foaming stability (%)

pH

5 A

zU/g

10m

in25

AzU

/g 2

0 m

in40

AzU

/g 4

0min

80 A

zU/g

120

min

SPI

G

H

143

Tab

le 4

.5

OR

AC

valu

es (

µm

ol T

E/g

) o

f se

lect

ed c

atfi

sh b

y-p

roduct

s hydro

lysa

tes

hydro

lyzed

under

dif

fere

nt

cond

itio

ns.

Valu

es

wer

e ex

pre

ssed

as

mea

n (

n=

3),

and s

tandar

d d

evia

tio

n w

as n

ot

over

10%

of

each m

ean v

alu

e;

valu

es

wit

hin

eac

h t

yp

e o

f

sam

ple

mar

ked

by t

he

dif

fere

nt

lett

er w

ithin

sam

e co

lum

n a

re s

ignif

icantl

y d

iffe

rent

(P <

0.0

5).

Pro

tease

s

5 A

zU/g

10

min

2

5 A

zU/g

20

min

4

0 A

zU/g

40

min

8

0 A

zU/g

12

0m

in

30

°C

40

°C

50

°C

60

°C

30

°C

40

°C

50

°C

60

°C

30

°C

40

°C

50

°C

60

°C

30

°C

40

°C

50

°C

60

°C

Fic

in

18

90

.5a

19

57

.8a

18

47

.2a

17

46

.2b

19

27

.3a

19

89

.4a

18

96

.2a

18

68

.4b

19

66

.1a

20

47

.7a

20

18

.4a

20

79

.4a

19

87

.3a

20

11

.5a

21

14

.5a

22

04

.1a

Pap

ain

1

28

3.2

c 1

30

1.5

c 1

35

5.1

c 1

39

7.5

c 1

31

1.2

c 1

37

8.4

c 1

39

7.4

b 1

53

2.7

c 1

47

7.2

c 1

42

7.9

c 1

50

2.7

c 1

59

5.3

c 1

50

4.3

b 1

46

6.4

c 1

55

4.6

c 1

62

2.7

c

Bro

mel

ain

1

83

1.8

b 1

79

4.1

b 1

80

5.8

b 1

87

8.4

a 1

87

2.5

b 1

85

5.3

b 1

86

1.7

a 1

90

1.2

a 1

92

2.5

b 1

89

3.3

b 1

95

5.3

b 1

97

3.2

b 1

97

3.6

a 1

90

9.5

b 2

04

1.4

b 2

01

5.8

b

Neu

trase

2

18

.3de

27

7.5

d 2

44

.8d

20

2.7

e 2

53

.6f

30

2.7

d 3

87

.4c

22

1.7

d 4

07

.1f

46

9.2

d 4

62

.3d

39

6.4

d 4

42

.4e

49

4.2

d 4

83

.1d

43

7.7

e

Alc

alase

2

27

.3d

17

6.3

f 1

85

.9e

17

9.6

f 3

55

.8d

20

1.7

g 2

23

.7de

24

1.9

d 7

58

.3d

31

5.6

e 3

87

.4e

29

6.5

e 8

73

.5c

33

2.8

f 4

39

.4e

45

1.2

d

Nov

o-p

roD

2

27

.4d

24

1.6

e 1

89

.6e

24

7.5

d 2

78

.9e

27

6.3

e 2

06

.9e

23

1.9

d 4

27

.5e

30

1.5

f 2

57

.3g

22

8.1

f 7

21

.7d

34

6.7

f 3

05

.7f

37

9.7

g

Pro

tam

ex

21

7.5

e 1

57

.9g

18

8.5

e 2

05

.9e

24

7.4

f 2

65

.9f

25

5.9

d 2

23

.8d

27

1.9

g 3

27

.5e

32

6.3

f 2

89

.4e

45

3.8

e 4

67

.3e

43

7.5

e 3

99

.6f

4

0°C

5

0°C

6

0°C

7

0°C

4

0°C

5

0°C

6

0°C

7

0°C

4

0°C

5

0°C

6

0°C

7

0°C

4

0°C

5

0°C

6

0°C

7

0°C

Ther

moly

sin

2

69

.5

25

7.4

2

66

.9

25

8.4

2

81

.6

32

7.7

3

76

.3

31

8.5

3

14

.8

33

7.7

3

96

.4

36

5.2

3

37

.5

35

7.9

4

21

.8

75

6.8

144

Tab

le 4

.6

DP

PH

valu

es (

µm

ol T

E/g

) o

f se

lect

ed c

atfi

sh b

y-p

roduct

s hydro

lysa

tes

hydro

lyzed

under

dif

fere

nt

cond

itio

ns.

Valu

es

wer

e ex

pre

ssed

as

mea

n (

n=

3),

and s

tandar

d d

evia

tio

n w

as n

ot

over

10%

of

each m

ean v

alu

e;

valu

es

wit

hin

eac

h t

yp

e o

f

sam

ple

mar

ked

by t

he

dif

fere

nt

lett

er w

ithin

sam

e co

lum

n a

re s

ignif

icantl

y d

iffe

rent

(P <

0.0

5).

Pro

teas

es

5 A

zU/g

10m

in

25

AzU

/g 2

0m

in

40

AzU

/g 4

0m

in

80 A

zU/g

120m

in

30°C

40°C

50°C

6

0°C

3

0°C

4

0°C

5

0°C

6

0°C

3

0°C

4

0°C

50°C

60°C

30°C

40°C

50°C

60°C

Fic

in

45.2

a 45.6

a 47.6

a 4

8.2

b

45

.8b

58

.3b

46

.1b

58

.3b

46

.1b

57

.6b

58.3

b

59.2

b

67.8

b

61.5

b

67.8

b

75.6

a

Pap

ain

33.3

c 38.3

b

34.3

c 3

7.3

c 3

1.7

c 3

8.6

c 3

3.0

c 4

5.5

c 3

8.6

c 4

2.9

c 47.8

c 48.4

c 42.5

c 44.8

c 46.8

c 55.6

b

Bro

mel

ain

39.2

b

46.1

a 42.9

b

54

.3a

51

.7a

55

.3a

54

.3a

69

.7a

74

.7a

66

.8a

65.8

a 69.7

a 69.4

a 71.4

a 72.7

a 74.3

a

Neu

tras

e 4.2

e 4.8

d

4.6

d

4.0

e 4

.7e

5.2

d

5.8

d

4.0

f 5

.8e

5.4

d

5.7

d

5.1

d

5.5

f 5.7

d

4.7

f 5.3

c

Alc

alas

e 4.8

d

3.7

e 3.8

e 3

.7f

5.4

d

3.8

f 4

.2f

4.5

d

8.1

d

4.3

f 5.2

e 4.8

e 9.0

d

4.7

g

5.6

e 5.4

c

No

vo-p

roD

4.6

e 4.9

c 3.8

e 4

.8d

4.8

e 5

.3d

3.7

g

4.0

f 6

.0e

4.4

f 4.5

f 4.0

f 7.1

e 5.1

f 4.2

g

4.7

d

Pro

tam

ex

4.2

e 3.8

e 3.5

f 3

.8ef

4.8

e 5

.0e

4.6

e 4

.2e

3.8

f 4

.7e

4.5

f 3.9

g

5.6

f 5.6

e 5.7

d

5.4

c

40°C

50°C

60°C

7

0°C

4

0°C

5

0°C

6

0°C

7

0°C

4

0°C

5

0°C

60°C

70°C

40°C

50°C

60°C

70°C

Th

erm

oly

sin

4.7

4.5

4.9

4

.7

5.1

5

.5

5.7

5

.0

5.3

5

.5

5.8

5.7

5.2

5.8

6.0

8.5

145

4.4.10 ACE inhibition assay

Angiotensin converting enzyme (ACE), is a central component of the renin-

angiotensin system (RAS), which controls blood pressure by regulating the volume of

fluids in the body. It converts the hormone angiotensin I to the active vasoconstrictor

angiotensin II. Therefore, ACE indirectly increases blood pressure by causing blood

vessels to constrict. ACE inhibitors are widely used as pharmaceutical drugs for

treatment of hypertension.

The IC50 values of selected hydrolysates are shown in Table 4.7. Generally,

hydrolysates with higher degree of hydrolysis (> 50%) exhibited stronger ACE inhibition

capacity than other hydrolysates with lower degree of hydrolysis (< 25%). No ACE

inhibition capacity was detected in hydrolysates with degree of hydrolysis lower than

30% when the concentration of hydrolysates over 2 mg/mL. In addition, positive control

captopril (1.11 ng/mL) showed stronger ACE inhibition capacity than the selected

hydrolysates. More recent, ACE inhibition capacity of leatherjacket muscle hydrolysates

has been reported with IC50 up to 7 mg/mL (Salampessy, Reddy, Phillips, &

Kailasapathy, 2017). Sardine by-product hydrolysates showed ACE inhibition capacity

with IC50 values up to 7.4 mg/mL (Bougatef et al., 2008). However, no positive control

was reported in the above literature and make comparison difficult. No literature has been

reported the hydrolysis of catfish by-products by using various proteases. We are the first

to report the hydrolysis of catfish by-products by using various proteases and set the

proteolytic activity to the same level.

No significant correlation between ORAC values and ACE inhibition was

observed. This might be because that the mechanisms of ACE inhibition and ORAC

146

assay was not the same. The ACE inhibition activity of peanut protein hydrolysates

increased with the increase of degree of hydrolysis (Jamdar et al., 2010). Therefore,

correlation between ACE inhibition and degree of hydrolysis was studied. A negative

correlation between degree of hydrolysis and ACE inhibition activity was observed with r

equal to 0.455 at 0.05 significant level. Sangsawad and coworkers have reported that

smaller molecular size of peptides possessed stronger ACE inhibitory activity

(Sangsawad et al., 2018). Regarding to the relationship between peptides structure and

ACE inhibitory activity, peptides had Pro, Phe, or Tyr at the C-terminus, and Val and Ile

at the N-terminus exhibited potent inhibitory activity against ACE (Tsai, Chen & Pan,

2008).

4.4.11 DPP-IV inhibition assay

DPP-IV plays a critical role in maintaining glucose homeostasis. It is responsible

for inactivating incretins, such as glucose-dependent insulinotropic polypeptide (GIP) and

glucagon-like peptide-1 (GLP-1). Secretion of these two intestinal hormones is triggered

by food intake. After DPP-IV was blocked by inhibitors, the increased incretin levels will

inhibit glucagon release, which in turn increases insulin secretion, decreases gastric

emptying, and decreases blood glucose levels.

IC50 values of the selected hydrolysates against DPP-IV is shown in Table 4.8.

The hydrolysates hydrolyzed by bacteria derived proteases showed no inhibition activity

against DPP-IV. However, hydrolysates hydrolyzed by plant derived proteases exhibited

inhibition activity against DPP-IV. The reason might be that the hydrolysates hydrolyzed

by papain, ficin and bromelain (plant derived proteases) had higher degree of hydrolysis

which resulted in smaller size of peptides. We are the first to report the DPP-IV inhibition

147

activity of catfish by-product hydrolysates. Therefore, comparison to the literature is

difficult.

However, two peptides Gly-Pro-Ala-Glu (372.4 Da) and Gly-Pro-Gly-Ala (300.4

Da) have been purified from Atlantic salmon skin gelatin hydrolysates and exhibited

potent DPP-IV inhibition activity (IC50 = 49.6 and 41.9 μM, respectively) (Li-Chan,

Hunag, Jao, Ho & Hsu, 2012). Barbel muscle hydrolysate has been separated by HPLC

into 5 fractions and two of them showed DPP-IV inhibition activity with IC50 = 1.09 and

0.21 mg/mL, respectively. Two peptides Trp-Ser-Gly and Phe-Ser-Asp have been

purified from the two fractions, and molecular docking analysis indicated that Trp-Ser-

Gly had stronger affinity for DPP-IV (Sila et al., 2016). Whey protein hydrolysates have

exhibited DPP-IV inhibition activity, and peptide Ile-Pro-Ala is a moderate DPP-IV

inhibitor (Tulipano, Sibilia, Caroli & Cocchi, 2011). Peptide (YINQMPQKSRE) purified

from egg yolk protein hydrolysates exhibited DPP-IV inhibition activity with IC50 value

of 222.8 μg/mL (Zambrowicz et al., 2015). Peptides length did not correlate with DPP-IV

inhibitory activity since peptides larger than 13 residues exhibited potent DPP-IV

inhibitory activity (Uchida, Ohshiba & Mogami, 2011). Amino acid composition is also

not the primary factor for DPP-IV inhibitory activity, because that the differences in

inhibitory activity between reverse and forward dipeptides Trp-Val was reported

(Nongonierma & FitzGerald, 2013). Further studies are needed to understand the

mechanism of DPP-IV inhibition activity produced by peptides.

148

Tab

le 4

.7

IC5

0 v

alu

es

(mg/m

L)

of

catf

ish b

y-p

roduct

s hydro

lysa

tes

hydro

lyzed

under

dif

fere

nt

cond

itio

ns

again

st a

ng

iote

nsi

n-

conver

ting e

nzym

e.

Valu

es

wer

e ex

pre

ssed

as

mea

n (

n=

3),

and s

tandar

d d

evia

tio

n w

as n

ot

over

12%

of

each m

ean v

alu

e. N

D =

not

det

ecta

ble

at

level

of

2 m

g/m

L.

Cap

topri

l w

as

use

d a

s po

siti

ve

contr

ol.

Pro

tease

s 5

AzU

/g 1

0m

in

25 A

zU

/g 2

0m

in

40 A

zU

/g 4

0m

in

80

AzU

/g 1

20

min

3

0°C

4

0°C

5

0°C

60°C

30°C

40°C

50°C

60°C

30°C

40°C

5

0°C

6

0°C

3

0°C

4

0°C

5

0°C

6

0°C

Fic

in

0.5

2b

1.1

5a

0.6

4c

0.7

4b

0.3

9b

0.4

0b

0.4

3b

0.3

8b

0.3

7a

0.3

3b

0.3

5b

0.3

0c

0.2

4d

0.1

7c

0.2

2b

0.2

1b

Papain

0

.47

c 0

.67

b

1.1

9a

0.5

5c

0.3

1c

0.5

7a

0.5

6 a

0.5

2a

0.3

3b

0.4

3a

0.4

2a

0.4

8a

0.2

3d

0.4

2a

0.3

0a

0.2

1b

Bro

mel

ain

0

.61

a 0

.57

c 0

.87

b

0.9

6a

0.4

6a

0.3

8c

0.3

9 c

0.3

9b

0.3

6a

0.2

6c

0.3

3b

c 0

.40

b

0.3

2c

0.2

8b

0.1

8c

0.1

8c

Neu

trase

N

D

ND

N

D

ND

N

D

ND

N

D

ND

N

D

ND

N

D

ND

N

D

ND

N

D

ND

Alc

ala

se

ND

N

D

ND

N

D

ND

N

D

ND

N

D

1.8

7

ND

N

D

ND

1

.84

b

ND

N

D

ND

Novo-

pro

D

ND

N

D

ND

N

D

ND

N

D

ND

N

D

ND

N

D

ND

N

D

1.9

5a

ND

N

D

ND

Pro

tam

ex

ND

N

D

ND

N

D

ND

N

D

ND

N

D

ND

N

D

ND

N

D

ND

N

D

ND

N

D

4

0°C

5

0°C

6

0°C

70°C

40°C

50°C

60°C

70°C

40°C

50°C

6

0°C

7

0°C

4

0°C

5

0°C

6

0°C

7

0°C

Ther

moly

s

in

ND

N

D

ND

N

D

ND

N

D

ND

N

D

ND

N

D

ND

N

D

ND

N

D

ND

1

.87

a

Capto

pri

l 1.1

1 (

ng/m

L)

149

Tab

le 4

.8

IC5

0 v

alu

es

(mg/m

L)

of

sele

cted

cat

fish

by

-pro

duct

s hydro

lysa

tes

hydro

lyze

d u

nder

dif

fere

nt

cond

itio

ns

again

st D

PP

-

IV.

Valu

es

wer

e ex

pre

ssed

as

mea

n (

n=

3),

and s

tandar

d d

evia

tio

n w

as n

ot

over

10%

of

each m

ean v

alu

e;

valu

es

wit

hin

eac

h t

yp

e o

f

sam

ple

mar

ked

by t

he

dif

fere

nt

lett

er w

ithin

sam

e co

lum

n a

re s

ignif

icantl

y d

iffe

rent

(P <

0.0

5).

Pro

teas

es

5 A

zU/g

10m

in

25 A

zU/g

20m

in

40 A

zU/g

40m

in

80 A

zU/g

120m

in

30°C

40°C

50°C

60°C

30°C

40°C

50°C

60°C

30°C

40°C

50°C

60°C

30°C

40°C

50°C

60°C

Fic

in

9.2

b

8.9

a 8.8

b

8.7

b

8.3

b

8.2

c 8.1

b

7.9

c 7.9

a 7.7

a 6.5

c 6.8

b

6.5

b

5.7

b

5.2

b

5.1

a

Pap

ain

9.1

b

8.5

c 9.1

a 8.6

b

8.4

b

9.1

a 9.2

a 8.9

b

7.5

b

7.2

b

7.8

a 7.5

a 6.7

a 4.8

c 5.4

a 5.2

a

Bro

mel

ain

9.4

a 8.7

b

9.2

a 9.7

a 8.7

a 8.4

b

8.1

b

9.0

a 7.3

c 6.9

c 7.2

b

6.4

c 6.3

c 6.2

a 5.4

a 4.9

b

Neu

tras

e N

D

ND

N

D

ND

N

D

ND

N

D

ND

1.8

7

ND

N

D

ND

N

D

ND

N

D

ND

Alc

alas

e N

D

ND

N

D

ND

N

D

ND

N

D

ND

N

D

ND

N

D

ND

N

D

ND

N

D

ND

Novo-p

roD

N

D

ND

N

D

ND

N

D

ND

N

D

ND

N

D

ND

N

D

ND

N

D

ND

N

D

ND

Pro

tam

ex

ND

N

D

ND

N

D

ND

N

D

ND

N

D

ND

N

D

ND

N

D

ND

N

D

ND

N

D

40°C

50°C

60°C

70°C

40°C

50°C

60°C

70°C

40°C

50°C

60°C

70°C

40°C

50°C

60°C

70°C

Ther

moly

sin

N

D

ND

N

D

ND

N

D

ND

N

D

ND

N

D

ND

N

D

ND

N

D

ND

N

D

ND

Sit

agli

pti

n

7.3

μg/m

L

150

4.4.12 Protein recovery rate of hydrolysates with good functionalities

Protein recovery rate of hydrolysates with good functionalities were analyzed, and

the results are shown in Table 4.9. The protein recovery rate of hydrolysates with good

functionalities were ranging from 35.9% to 59.7%. In general, papain hydrolysates

showed higher protein recovery rate than others. Capelin protein hydrolysates were

obtained by Alcalase, neutrase and papain hydrolysis, and the protein recovery rate was

ranging from 51.6% to 70.6%, and Alcalase hydrolysates exhibited the highest protein

recovery rate (70.6%) (Shahidi, Han, & Synowiecki, 1995). However, the enzyme was

added based on the weight basis instead of enzyme proteolytic activity. Therefore, it is

impossible to compare the protein recovery rate from the present study with literature at

the same enzyme proteolytic activity. In addition, the present study is the first to report

the Novo-proD hydrolysate obtained from fish proteins, thus make comparison difficult.

Protein hydrolysates were obtained from soybean hulls with Novo-proD (Rojas, Siqueira,

iranda, Tardioli & Giordano, 2014). However, Novo-proD was added on the weight basis

which make comparison impossible.

Comparing the yield of enzymatic hydrolysis to the alkaline extraction (Chapter

5) of the proteins were extracted with different pH’s ranging from 7.5 to 11 at 0.5

interval, and the results indicated that protein recovery rate was the highest (42.8%) from

pH 8.5 extraction and without compromising the protein integrity. The proteins yield was

36.19% (wet basis) for pH 8.5 extraction, and moisture of the extracted protein is around

78%, and the protein content of the extracted protein is 83.71% (dry weight basis). In

addition, the protein content of mixture is 14.85% (wet weight basis). Therefore, the

protein recovery rate of the pH 8.5 extraction was = [0.3619 * (1-0.78) * 0.8371]

151

/(1*0.1485)=42.84%. Enzymatic hydrolysis is an alternative method to utilize the by-

product in terms of the protein recovery rate. However, the bitterness of hydrolysates

should be masked or eliminated if applied to human foods.

4.5 Conclusion

Ficin (80 AzU/g) was the most efficient in hydrolyzing the ground catfish by-

product (DH reaching 71%) in 120 min at 30°C among all the enzymes. The hydrolysis

curves fit the Peleg model well with R2 higher than 0.91. Ficin (80 AzU/g) was the most

efficient in hydrolyzing the ground catfish by-product (DH reaching 71.88%) in 120 min

at 30°C among all the enzymes. Thermolysin could be used for industries to hydrolyze

protein by-products in terms of hydrolysis efficiency and economy. Functional properties

of hydrolysates were affected by pH, and hydrolysates exhibited potent emulsion and

may be used as a replacement of SPI in food processing.

152

Tab

le 4

.9

Pro

tein

rec

over

y r

ate

(%)

of

hydro

lysa

tes

wit

h g

ood f

unct

ionali

ties

.

Valu

es

wer

e ex

pre

ssed

as

mea

n (

n=

3),

and s

tandar

d d

evia

tio

n w

as n

ot

over

10%

of

each m

ean v

alu

e;

valu

es

wit

hin

eac

h t

yp

e o

f

sam

ple

mar

ked

by t

he

dif

fere

nt

lett

er w

ithin

sam

e co

lum

n a

re s

ignif

icantl

y d

iffe

rent

(P <

0.0

5).

Pro

tease

s

5 A

zU/g

10

min

2

5 A

zU/g

20

min

4

0 A

zU/g

40

min

8

0 A

zU/g

12

0m

in

30

°C

40

°C

50

°C

60

°C

30

°C

40

°C

50

°C

60

°C

30

°C

40

°C

50

°C

60

°C

30

°C

40

°C

50

°C

60

°C

Fic

in

43

.7%

c 4

5.6

%b

45

.0%

b

44

.2%

b

44

.9%

b

46

.5%

b

46

.1%

b

43

.9%

b

43

.1%

c 4

3.1

%b

45

.8%

b

44

.4%

b

44

.7%

b

42

.3%

b

47

.1%

b

43

.5%

b

Pap

ain

5

6.7

%a

59

.7%

a 5

5.2

%a

56

.2%

a 5

5.7

%a

58

.4%

a 5

5.7

%a

54

.8%

a 5

7.9

%a

58

.4%

a 5

5.4

%a

55

.7%

a 6

0.2

%a

59

.4%

a 5

5.5

%a

54

.3%

a

Bro

mel

ain

4

1.1

%d

42

.0%

c 3

9.2

%c

44

.5%

b

39

.7%

c 4

0.5

%c

38

.3%

c 4

3.8

%b

38

.5%

d

39

.4%

b

35

.9%

c 3

7.1

%c

39

.9%

c 3

9.0

%c

34

.9%

c 3

6.2

%c

Nov

o-

pro

D

48

.9%

b

NA

4

6.8

%b

NA

4

8.4

%b

NA

4

7.3

%b

NA

4

8.8

%b

NA

4

6.8

%b

NA

N

A

NA

4

8.4

%b

NA

4

0°C

5

0°C

6

0°C

7

0°C

4

0°C

5

0°C

6

0°C

7

0°C

4

0°C

5

0°C

6

0°C

7

0°C

4

0°C

5

0°C

6

0°C

7

0°C

Ther

moly

sin

3

6.3

%

39

.3%

4

2.2

%

51

.2%

3

7.1

%

40

.2%

4

1.6

%

49

.3%

3

6.1

%

39

.9%

4

2.0

%

47

.7%

3

5.2

%

40

.1%

4

1.3

%

44

.8%

153

4.6 Research diagram in proposal

Research diagram of hydrolyzing catfish by-products in proposal are shown in

Figure 4.18. Future work and limitation of the current study will be discussed in Chapter

VI.

Figure 4.18 Research diagram of utilization of catfish by-product for making protein

hydrolysates.

154

CHAPTER V

EFFECT OF EXTRACTION PH AND TRANSGLUTAMINASE CROSS-LINKING

ON PHYSICOCHEMICAL AND TEXTURAL CHANGES IN MYOFIBRILLAR

PROTEINS EXTRACTED FROM CHANNEL CATFISH BY-PRODUCTS

5.1 Abstract

Myofibrillar proteins were firstly extracted from catfish by-products (mixture of

heads and frames) with different pH conditions, and then made into protein gels.

Transglutaminase (TGase) was incorporated to improve the gel structure. Solubility and

secondary structure of extracted myofibrillar proteins were studied. Gelling properties of

the protein gels were studied by dynamic rheological measurement. Physicochemical,

textural and thermal properties of protein gels treated with TGase were investigated.

Protein pattern changes of TGase treated protein gel was studied by SDS-PAGE. Results

indicated that alpha-helicity of myofibrillar proteins decreased with the extraction pH

over 9. Storage modulus (G’) of protein sol decreased as the increase of extraction pH (9-

11). Transglutaminase treatment had significant effect on the denaturation temperature

and enthalpy of protein gel. Rheological measurement of TGase treated protein sol

showed that excessive TGase could weaken the gel structure. Myofirbillar proteins could

extracted from catfish by-products and made into protein gels which is a value-added

product.

155

Keywords: Channel catfish by-products, circular dichroism, dynamic rheology,

differential scanning calorimetry, transglutaminase, protein gel.

5.2 Introduction

Channel catfish aquaculture is one of the most important aquaculture in United

State. Mississippi state ranks No. 1 in catfish production for continuous 5 years (Catfish

production (USDA, 2013, 2014, 2015, 2016 and 2017). Channel catfish fillet production

is one of the most important part of the local economy. However, large number of by-

products were generated during fillet processing which including skin, frames and heads.

Collagen has been successfully extracted from catfish skin with high yield and reported

earlier (Tan & Chang, 2018). Those by-products account more than 50% of the whole

fish weight which were disposed. Those by-products were regarded as waste and most of

them were processed into aquaculture feed, otherwise, it would cause serious

environment pollution. In addition, there is a shortage of protein supply worldwide, and

recovering proteins from catfish by-products is an excellent way to provide low-cost

animal proteins to fulfill global human nutritional needs.

Traditionally, surimi is made from fish meat of both salt and fresh water origins.

China is the first to produce fish ball back to year 220 BC (Park, Nozaki, Suzuki, &

Beliveau, 2013). In United State, crabsticks are made from Alaska Pollock or Pacific

Whiting after extensive washing. However, Channel catfish has never been used

commercially to make surimi or surimi like products since it has higher value in the form

of fillet. Using fish meat to make surimi products such as fish ball, fish cake and fish tofu

have been studied thoroughly (Jenkelunas & Chan, 2018; Moon, Yoon, & Park, 2017).

Protein isolate from catfish fillet have been extracted by acid and alkaline extraction

156

methods and made into protein gels (Kristinsson & Liang, 2006). However, such protein

gel made from catfish fillet has remained only in the stage of research since using fillet to

make protein gel is cost prohibitive. In addition, the nativity of the proteins is affected by

extraction/processing technology that imposes shearing force during grinding and mixing,

and degradation by endogenous hydrolytic enzymes that need to be controlled by cooling

and use of protease inhibitors. It is a challenge to maximize processing yield and the

quality of protein end-products from less desirable raw materials such as fish by-

products. Research for making food-grade products from the by-products is still in its

infancy. And this is the first study to show that the possibility of using proteins extracted

from catfish by-products (mixture of heads and frames) to make protein gel (surimi like

product), and to compare secondary structure, physicochemical and thermal properties of

proteins under different alkaline extraction conditions. The objective of this study was to

investigate (1) the secondary structure changes of myofibrillar proteins under different

extraction pHs, (2) physicochemical, textural and thermal properties of the protein gel

made from proteins extracted from catfish by-products, and (3) the effect of

transglutaminase on protein gels.

5.3 Materials and methods

5.3.1 Materials

Catfish heads and frames were collected from local catfish fillet processing

company (Country Select, Isola, MS). All the samples were buried in crushed ice during

transportation to laboratory. Heads and frames were ground using a meat grinder (LEM

Products #32, West Chester, OH) for six cycles, and the ground frames and heads were

mixed with 3% sorbitol, 3% polytriphosphate and 3% sucrose, and stored at -80°C until

157

use.

5.3.2 Chemicals

All chemicals, reagents were analytical grade and obtained from Sigma-Aldrich

Chemical Company (St. Louis, MO, USA). Sucrose and sorbitol are food grade and

obtained from local market.

5.3.3 Myofibrillar protein extraction

Ground catfish heads and frames were well mixed (w/w = 3:2). The by-products

mixture was mixed with 4 times volume of chilled water which contained 0.5% EDTA.

The pH of the slurry was adjusted to 7.5, 8, 8.5, 9, 9.5, 10, 10.5 and 11, and extract for 30

min at 4 °C with continuous stirring. The slurry was then passed through 0.5 mm sieve to

remove the pigment, and the pH of the protein solution was adjusted to 5.5 and

centrifuged at 5000 rpm for 30 min at 4 °C. Discard the supernatant and oil layer, weight

the pallet and measure the moisture content of pallet. Yield (%) = weight of pallet (g)/

weight of the mixed by-products (g). The precipitated proteins were collected for further

analysis and protein gel making.

5.3.4 Circular dichroism (CD)

Extracted myofibrillar proteins were dissolved in 0.3 M NaCl and diluted to a

concentration of 0.3 mg/mL. CD spectrum were recorded by a Jasco 810

spectropolarimeter (Jasco International Co., Easton, MD) with a Peltier jacketed

software-controlled multiple cuvette holder according to our previous study (Tan &

Chang, 2018). Myofibrillar proteins extracted with pH 7 was used as control. CD

spectrum of the myofibrillar proteins were analyzed using a 0.1 cm path length quartz

158

cell for far-UV measurement. Spectrum was recorded from 200 to 250 nm, and the scan

rate was set as 100 nm/min. Five scans were taken for each sample. Estimate α-helicity

was calculated according to the previous method (Liu, Zhao, Xiong, Xie, & Liu, 2007). A

mean residue weight of 110 g/mol has been assumed.

5.3.5 Effect of pH on solubility of extracted myofibrillar proteins

The relative solubility test was determined according to the previous method with

some modifications (Kristinsson, Theodore, Demir, & Ingadottir, 2005). Extracted

myofibrillar proteins were mixed with deionized water (1:5) and adjust pH to

corresponding extraction pH, then homogenized for 1 min (6000 rpm). Determine the

protein content in the protein homogenate with Bradford assay (Bradford, 1976), and

adjust the final protein concentration to 10 mg/mL. Then separate the protein solution to

12 fractions, adjust the pH of each fraction from 1 to 12 at 1 pH interval. Centrifuge to

precipitate the undissolved proteins, the protein concentration in supernatant was

determined by Bradford assay (Bradford, 1976). The relative protein solubility was

calculated by dividing the protein content in supernatant by the protein content before pH

adjustment.

5.3.6 Production of protein gel and transglutaminase treatment

Protein gel preparation was performed according to the method reported

previously (Huang, Abdel-Aal, & Awad, 1996). The pH of extracted myofibrillar proteins

was adjusted to neutral, and the moisture content was adjusted to 80% by adding water.

Then mixed with 3 % NaCl and 3% potato starch by grinding using a mortar and pestle

for 100 times (clockwise) at room temperature to obtain protein sol. Transglutaminase at

159

the concentration of 0.1, 0.5, 1, 2 and 4 U/g of proteins (protein content was determined

by Kjeldahl method) were added to proteins extracted with pH 8.5 before mixing by

using mortar and pestle. Protein sol treated with 1 U/g of TGase was incubated for

different times (0-120 min) at 40°C to study the protein changes during TGase treatment.

The well mixed protein sol was split into two parts, one part was stuffed into 60 mL

syringe (3 cm in diameter and 10 cm in length), and the other parts were saved for

dynamic rheological measurement. The syringes were placed in a stainless-steel rack and

covered with a glass ball on the top to prevent water evaporation during incubation. The

rack was then incubated at 40 °C water bath for 30 min and then 90°C for 20 min. Rack

was taken out after time achieved, and store at 4°C overnight to obtain protein gel.

5.3.7 Dynamic rheological measurement

The well mixed protein sol made from myofibrillar proteins extracted under

different pH were used for dynamic rheological measurement. Dynamic rheological

measurements were performed on an Anton paar dynamic rheometer (MCR 502, Anton

Paar, Houston, TX, USA). A parallel-plate geometry of 25 mm diameter and normal

force of 1 N were set for measurement. The protein sol was placed between the parallel

plates of the rheometer. The excess samples protruding beyond the upper plate was

removed. Silicone oil was gently applied to the edge of exposed sample to prevent

moisture loss during temperature increasing. Protein sol on the parallel plates were

allowed to rest for 2 min before analysis to ensure both thermal and mechanical

equilibrium at the time of measurement. For temperature sweep, samples were heated at a

speed of 1°C/min from 10 to 90 °C with 0.1% shear strain (γ) and 10 Hz frequency.

160

Storage modulus (G’) was recorded from the temperature sweep test. Three replicates of

each sample were measured.

5.3.8 Differential scanning calorimetry (DSC)

Myofibrillar proteins extracted with pH 8.5 was made into protein sol and treated

with transglutaminase at different concentrations. The protein gels were used for

differential scanning calorimetry analysis. DSC was carried out using a differential

scanning calorimetry (Perkin Elmer, Model DSC8000, Norwalk, CA, USA). Twenty

micrograms of cooked protein gels were exactly weighted and transferred to aluminum

pan and sealed, and empty pan was used as reference. The sample was scanned at

1.5°C/min over the range of 20-80°C in a nitrogen atmosphere.

5.3.9 Textural analysis and cooking loss

Textural analysis of the protein gel was carried out by using TA.XTplus Texture

Analyzer (Stable Microsystems, Godalming, Surrey, UK). Protein gels were removed

from the syringe and cut into specimens with 2.5 cm in height. Breaking force (g) and

deformation distance (mm) were measured by texture analyzer which equipped with a 5-

mm diameter plunger and the depression speed was set at 60 mm/min. The syringe and

protein gels were weighed before and after cooking. Cooking loss (%) was calculated by

dividing weight difference before and after cooking by weight before cooking.

5.3.10 Color

The color of protein gels was measured by a benchtop color meter

(Hunter Lab Colorimeter D-25, Hunter Associates Laboratory, Ruston, USA). Hunter

color (“L”, “a”, “b”) of protein gels were measured according to our previous method

161

(Kong, Chang, Liu, & Wilson, 2008). Three random spots of each sample were measured

and the averages were recorded.

5.3.11 SDS-PAGE analysis

Electrophoresis was carried out according to the method of Laemmli (1970).

Protein profiles of extracted myofibrillar protein and proteins during each extraction

stage were analyzed by SDS-PAGE gel, protein pattern changes upon TGase treatment (1

U/g of proteins, incubated at 40°C for 0-120 min) was studied as well. Electrophoresis

was performed on a 2-20% gradient gel (16×20 cm). Twenty-five micrograms of proteins

were loaded to each well, electrophoresis was conducted at room temperature with

circulating cooling water at 80V for 1.5 hrs, and 125V for 2 hrs, then 150V for 2hrs, and

followed 175V for 1 hr. Proteins were stained with 0.1% Commassie Brilliant Blue R-

250 dissolved in water, methanol and acetic acid (9:9:2, v/v/v) for 1 hr, then destained

using a solution containing water, methanol and acetic acid (17:1:2, v/v/v). For the

quantification of new polymers and other bands, gels were scanned and analyzed by a

Molecular Imager (Bio-Rad Chemidoc™ XRS+, Hercules, CA, USA) equipped with

Image Lab™Analysis Software (version 5.3).

5.3.12 Scanning electron microscopy (SEM)

Cooked protein gels (0.25×0.25×0.25 cm3) were fixed with 2.5% glutaraldehyde

in 0.2 M Sorenson’s phosphate buffer (pH 7.2) for 2 hrs at room temperature. Fixed

specimens were dehydrated in ethanol solution with serial concentrations of 30%, 50%,

70%, 85%, 95% and 100%, and then critical point dried (Autosamdri-931, Tousimis,

MD) using CO2 as transition fluid. The prepared samples were mounted on copper

162

specimen holders, sputter-coated with gold by Turbo-pumped sputter and carbon coater

(EMS 150T ES, Hatfield, PA). And examined on a JSM 6500F scanning electron

microscope (JEOL, Ltd., Akishima, Japan).

5.3.13 Statistical analysis

All the experiments were carried out on a completely randomized design and

performed in triplicate. Data were analyzed by ANOVA using 2016 SAS (version 9.4,

SAS Inc., Cary, NV, USA). Duncan’s multiple range test was carried out to determine

any significant differences between different pH extraction conditions.

5.4 Results and discussion

5.4.1 Yield of extracted myofibrillar proteins

Yield of myofibrillar proteins increased from 29.09% to 56.70% (wet weight

basis) when the extraction pH increases from 7.5 to 11. Strong correlation between

estimated α-helicity (%) of myofibrillar proteins extracted under different pH and yield

(moisture removed) was observed (R2 = 0.85). This observation showed a negative

correlation between yield and α-helicity of extracted proteins. The α-helicity of extracted

proteins decreased with the increase of extraction pH (9-11). Previous study had reported

that alkaline treatment decrease the α-helical structure of myosin extracted from catfish

fillet, however, only few pH points (pH 11, 11.5 and 12) were studied and no yield data

were reported (Raghavan & Kristinsson, 2008). We are the first to report the correlation

between protein yield and estimated α-helicity, which provide important information on

protein conformational changes upon alkaline extraction. Previous report indicated that if

a yield of 20% could be achieved, it would be economically feasible (McAlpin, Dillard,

163

Kim, & Montanez, 1994). Yield over 20% could be obtained in the current study without

compromise the α-helicity of the extracted proteins (36.19% with pH 8.5 extraction).

Therefore, to extract myofibrillar proteins from catfish by-products for value-added

product is practicable based on the economic analysis and the data obtained in the present

study.

5.4.2 Circular dichroism of extracted myofibrillar proteins

In order to gain insight into the secondary structure of extracted myofibrillar

proteins, secondary structure of extracted myofibrillar proteins was analyzed by circular

dichroism. CD spectrum of myofibrillar proteins extracted under different pH are shown

in Figure 5.1. Secondary structure of myofibrillar proteins was significantly affected by

extraction pH at the range of 9-11. The α-helical structure of proteins could be stabilized

by the hydrogen bond between carbonyl oxygen and amino hydrogen (Bastidas, Green,

Sprague, & Peters, 2016), and the electrostatic interaction between amino acids

(Adhikari, Somerset, Stull, & Fajer, 1999). CD spectrum of all the extracted proteins

exhibited two negative peaks at 222 and 208 nm, which is the predominant presence of α-

helical structure (Greenfield, 1999). CD spectrum of myofibrillar proteins extracted under

pH 7.5-8.5 were similar to control. However, the intensities of peaks were decreased with

extraction pH increasing (from 9-11), which indicated that the loss of α-helical

conformation upon alkaline extraction (pH 9-11). And relative α-helicity% of extracted

proteins was decreased from 36.6% to 28.8% with the extraction pH increased from 8.5

to 11. Conformational changes of myofibrillar proteins after alkaline treatment was

observed before (Raghavan & Kristinsson, 2008). The loss of α-helical structure of

myofibrillar proteins was also observed after the oxidation of proteins (Cao & Xiong,

164

2015). CD spectrum of myofibrillar proteins after treated with high pressure (200 to 600

MPa) exhibited significant lower peak intensity at 222 and 208 nm (Villamonte, Jury,

Jung, & Lamballerie, 2015). Most recent, the CD spectrum of fresh and thawed

myofibrillar proteins were reported, the results showed the loss of α-helical conformation

after thawing the meat in air (Jia, Nirasawa, Ji, Luo, & Liu, 2018). Though the loss of α-

helical structure of myofibrillar proteins had been reported, no study on gels which made

from those myofibrillar proteins has been reported. The present study was the first to

report the secondary structure changes of myofibrillar proteins extracted with broader

alkaline conditions (7.5-11). Therefore, this makes comparing the current results with

literature difficult.

165

Figure 5.1 Circular dichroism spectrum of myofibrillar proteins extracted under

different pH.

5.4.3 Effect of pH on solubility of extracted myofibrillar proteins

Myofibrillar proteins extracted under different pH were used for relative solubility

test. The effect of pH on solubility of extracted myofibrillar proteins are shown in Figure

5.2. The protein solubility followed a U shape curve, a typical of muscle protein

solubility curve (Choi & Park, 2002). The protein solubility sharply increased below pH

5 and above 10. The greatest solubility was observed at pH 3 in the acid range and pH 12

in the alkaline range. It was reported that the isoelectric point of myofibrillar proteins

from Channel catfish is 5.5 (Kristinsson, Theodore, Demir, & Ingadottir, 2005). When

the pH of protein solution close to the isoelectric point, the net charge of the protein

molecules was close to 0, which lead to protein precipitation. And this is the reason why

-25

-23

-21

-19

-17

-15

-13

-11

-9

-7

-5

200 205 210 215 220 225 230 235 240 245 250

CD

(m

deg

)

Wavelength (nm)

pH 7.5 pH 8 pH 8.5 pH 9 pH 9.5

pH 10 pH 10.5 pH 11 control

166

the extracted myofibrillar proteins exhibited extremely low solubility (ranging from

1.69% to 3.35%) at pH 5 and 6. In contrast, when the pH of protein solution was far away

from isoelectric point, the solubility of protein was increased by the repulsion forces (Tan

& Chang, 2018). Myofibrillar proteins extracted with weak alkaline conditions (pH 7.5-

8.5) exhibited higher solubility than that of higher pH (9-11) extracted proteins (pH 8.5

extracted proteins showed 13.73% higher solubility than the pH 11 extracted proteins at

pH 8). This observation was consistent with the CD spectra which indicated that the

myofibrillar proteins extracted at lower pH had similar CD spectra with control group

(pH 7 extraction). Loss of α-helical structure (partially denatured) would lead to

decreased solubility. Increased solubility at extreme pH conditions has been attributed to

the negative or positive charge of the protein at high or low pH, respectively.

167

Figure 5.2 Effect of pH on the solubility of myofibrillar proteins extracted under

different pH.

5.4.4 Rheological properties of protein gel made from extracted myofibrillar

proteins treated with and without transglutaminase

Traditionally, surimi was made from fish fillet with addition of salt (Park, Nozaki,

Suzuki, & Beliveau, 2013). Proteins extracted from fish muscle were used for protein gel

making recently (Xiong, et al., 2009; Zhang, Xu, Li, Wang, Xue, & Xue, 2018).

However, the current study was the first to report protein gel made from myofibrillar

proteins extracted from catfish by-products (the mixture of heads and frames). The

protein sol (before cooking) were made from myofibrillar proteins which extracted under

different pH, and the rheological profiles of those protein sol are shown in Figure 5.3.

Storage modulus (G’) representing elastic properties, it increased from 20 °C and reached

0.00%

10.00%

20.00%

30.00%

40.00%

50.00%

60.00%

70.00%

80.00%

90.00%

100.00%

0 1 2 3 4 5 6 7 8 9 10 11 12

Rel

ativ

e so

lubil

ity (

%)

pH

pH 7.5 pH 8 pH 8.5 pH 9

pH 9.5 pH 10 pH 10.5 pH 11

168

to the first peak at 43.47 °C for all the samples except protein sol made from pH 11

extracted myofibrillar proteins, suggesting the transformation from a viscous sol to an

elastic network (Liu, Zhao, Xiong, Xie, & Liu, 2007). And then the storage modulus G’

decreased reaching to the valley at around 48 °C. The decreasing in G’ might be caused

by the endogenous proteases in the protein sol, which lead to the dissociation of actin-

myosin and the denaturation of myosin tail. The first peak of protein sol made from pH

11 extracted myofibrillar protein was at 38.4 °C, which was lower than that of other

samples. As temperature further increasing, the second peak of protein sol made from pH

11 extracted proteins (56.63 °C) was much lower than others (about 73 °C). An

increasing in storage modulus (G′), which represents energy recovered per cycle of

sinusoidal shear deformation, indicated an increase in rigidity of the sample associated

with the formation of elastic gel structure (Yin, Reed, & Park, 2014). G’ of protein sol

made from pH 7.5, 8 and 8.5 extracted myofibrillar proteins were very similar. G’

decreased significantly (82.1% decreased at 90°C) with the increase of protein extraction

pH (8.5-11). This observation and the CD spectrum was mutual supportive. Correlation

analysis between storage modulus at the first peak and estimated α-helicity (%) was

conducted. The result indicated that the storage modulus (G’) of protein sol made from

different pH extracted myofibrillar proteins is positive correlated (R2 = 0.94) with α-

helicity (%) of the corresponding myofibrillar proteins. Myofibrillar proteins were

extracted from catfish muscle with alkaline conditions (pH 11, 11.5 and 12), and the

rheological properties of extracted proteins was studied (Raghavan & Kristinsson, 2008).

However, the rheological properties were determined on protein solution instead of

protein sol which make comparison difficult.

169

Transglutaminase was used to improve the gel properties of surimi (Cando,

Herranz, Borderías, & Moreno, 2016; Hu, et al., 2015). Transglutaminase could catalyze

the acyl transfer reaction, and creates larger molecules from small protein substrates

through the cross-linking reaction between glutamine residues and lysine residues in

protein molecules (Kuraishi, Sakamoto, & Soeda, 1996). The protein sol made from pH

8.5 extracted myofibrillar protein from catfish by-products was used for further TGase

treatment (1U/g of protein). Different concentrations of transglutaminase (TGase) were

mixed with extracted proteins and set for 2 hrs at 40 °C. The rheological profiles of those

TGase treated protein sols are shown in Figure 5.4. In general, storage modulus of protein

sol was increased as the addition of TGase. The storage modulus (G’) increase 42.1% at

10°C with the addition of TGase of 0 to 4 U/g of proteins. Storage modulus (G’)

representing elastic properties increased from 20 °C and reached the first peak at 43.47

°C for all the samples except protein sol treated with 2 and 4 U/g of proteins, which

indicated that higher concentrations of TGase could not help to improve the gel structure

of protein sol. In addition, storage modulus of protein sol treated with 2 and 4 U/g of

proteins were much lower than 1 U/g of proteins at 90 °C (19.9% and 41.0% lower than 1

U/g treated protein gels, respectively), which indicated that the denaturation of the

structure of protein sols. No literature has reported the rheological properties of protein

sol made from myofibrillar proteins extracted from catfish by-products, therefore,

comparison with literature was impossible.

170

Figure 5.3 Storage modulus (G’) of protein sol made from myofibrillar proteins

extracted under different pH.

0

2000

4000

6000

8000

10000

12000

14000

16000

18000

0 10 20 30 40 50 60 70 80 90 100

Sto

rage

Mo

du

lus

G' [

Pa]

Temperature (°C)

control pH 7.5 pH 8 pH 8.5 pH 9 pH 9.5 pH 10 pH 10.5 pH 11

171

Figure 5.4 Storage modulus (G’) of transglutaminase treated protein sol made from

myofibrillar proteins extracted from catfish by-products (mixture of frames

and heads).

Protein sol were treated with transglutaminase at 0 to 4 U/g of proteins level.

5.4.5 Differential scanning calorimetry

Enthalpy and maximum temperature of denaturation are two main factors in

determination of thermal properties of surimi and surimi-like products. The Tmax and ∆H

of protein sol treated with different concentrations of transglutaminase and control group

(without TGase) were recorded by DSC, the results are shown in Figure 5.5.

Thermograms of all samples showed two endothermic peaks, which are corresponding to

the thermal denaturation of myosin and actin, respectively (Wright, Leach, & Wilding,

1977). Results showed that transglutaminase treatment had significant effect on

0

2000

4000

6000

8000

10000

12000

14000

16000

18000

0 10 20 30 40 50 60 70 80 90 100

Sto

rage

Mo

du

lus

G' [

Pa]

Temperature (°C)

control 0.1 U/g 0.5 U/g 1 U/g 2 U/g 4 U/g

172

denaturation temperature and enthalpy of protein gels. Maximum temperature of

denaturation for myosin significantly increased from 43.13 to 45.29 °C when the

transglutaminase concentration was increased from 0 to 1 U/g of proteins. However,

maximum transition temperature decreased from 45.29 to 38.34°C for protein gel treated

with TGase of 2 and 4 U/g of proteins. The Tmax of actin treated with 1 U/g of proteins

was significant higher than that of control. There is no significant difference between the

enthalpy of myosin in protein gels treated with 2 and 4 U/g of proteins. A significant

increase in enthalpy of myosin was observed when transglutaminase concentration

increased from 0.1 to 1 U/g of proteins, which suggested that higher transglutaminase

concentration induced more denaturation and thus better texture characteristics (Park,

Cho, Kimura, Nozawa, & Seki, 2005). Transglutaminase could catalyze the formation of

iso-peptide bond between glutamine and lysine, and this iso-peptide bonds is important

for the formation of stable protein network. And this cross-linking results in the changes

in the hydrophobicity of the protein surface, and affecting the denaturation temperature of

proteins.

173

Figure 5.5 Differential scanning calorimetry thermograms of transglutaminase-treated

protein gels made from myofibrillar proteins extracted from catfish by-

products.

5.4.6 Penetration test, cooking loss and color of transglutaminase treated

protein gels

The punch test is the most popular measurement technique used in surimi industry

for evaluating the gel properties of surimi (Park, 2013). Breaking force, deformation

distance, cooking loss and color of TGase treated protein gel are shown in Table 5.1.

Breaking force and deformation distance of TGase treated protein gel were increased

with the addition of TGase from 0-1 U/g of proteins. However, breaking force and

deformation distance was decreased when the concentrations of TGase over 1 U/g of

20 30 40 50 60 70 80 90

Heat

flow

(W

/g)

Temperature (°C)

control 0.1 U/g 0.5 U/g 1 U/g 2 U/g 4 U/g

Tmax

= 43.13 °C, ∆H = 1.23 J/g

Tmax

= 76.58 °C, ∆H = 0.87 J/g

Tmax

= 43.37 °C, ∆H = 1.54 J/g T

max = 76.48 °C, ∆H = 0.95 J/g

Tmax

= 44.28 °C, ∆H = 2.04 J/g

Tmax

= 76.67 °C, ∆H = 1.15 J/g

Tmax

= 45.29 °C, ∆H = 2.21 J/g

Tmax

= 76.62 °C, ∆H = 1.22 J/g

Tmax

= 41.27 °C, ∆H = 1.48 J/g

Tmax

= 76.55 °C, ∆H = 0.84 J/g

Tmax

= 76.47 °C, ∆H = 0.81 J/g

Tmax

= 38.34 °C, ∆H = 1.43 J/g

174

proteins. This observation was consistent with previous report that breaking force and

deformation distance of surimi gel made from tilapia fillet would decreased when

excessive TGase was added (Seighalani, Bakar, Saari, & Khoddami, 2017). Protein gel

treated with 1 U/g of proteins showed the highest breaking force and deformation

distance. The cooking loss significantly increased when the TGase concentration over 1

U/g of proteins. The use of TGase above an optimum concentration caused a detrimental

effect on the textural property of surimi gels had been reported (Ramırez, Rodrıguez-

Sosa, Morales, & Vázquez, 2000). The above report was in accordance with the other

study that the breaking force and deformation of surimi gels made from Pollock increased

when TGase increased up to a certain level and further increase in TGase decreased the

gel strength of surimi gels (Jiang, Hsieh, Ho, & Chung, 2000). TGase induces the cross-

linking between glutamine and lysine residues through non-disulfide covalent bonds

(Zhang, Xu, Li, Wang, Xue, & Xue, 2018), which lead to the formation of myosin heavy

chain cross-linking and then result a stronger gel structure (Ma, Yi, Yu, Li, & Chen,

2015). Decreasing breaking force and deformation distance of TGase treated protein gel

may attributed to excessive cross-linking. The excessive cross-linking would decrease the

gel strength by impeding intermolecular aggregation, therefore, the network formation

would be affected (Hu, et al., 2018).

Cooking loss of protein gels was significantly affected by adding TGase. Cooking

loss was significantly lower than that of control by adding TGase at the level from 0.1 to

1 U/g of proteins, and increased when the concentrations of TGase over 1 U/g of

proteins. Most of the study had been reported the textural profiles of surimi or surimi-like

product but without reporting the cooking loss (Seighalani, Bakar, Saari, & Khoddami,

175

2017; Guo et al., 2017). And it has been reported that cooking loss of surimi gel was

decreased by adding TGase (Pietrasik, 2003). Increased cooking loss of TGase treated

surimi has not been reported, however, the increased expressible water of excessive

TGase treated surimi has been observed (Seighalani, Bakar, Saari, & Khoddami, 2017),

which suggesting that excessive cross-linking might weaken the water holding capacity

of surimi gel. Water holding capacity is an ability of proteins to soak up water and retain

it within a protein matrix. This kind of trapped water consist of trapped water,

hydrodynamic water and bounded water. The higher the TGase concentration, the greater

is the number of inter and intrachain peptide cross-linking and the lower the water-

protein interaction which results in higher cooking loss. In addition, the color of the

protein gels did not significantly affect by the addition of TGase.

176

Tab

le 5

.1

Tex

tura

l analy

sis,

co

okin

g l

oss

and c

olo

r o

f T

Gas

e tr

eate

d p

rote

in g

els

.

Dat

a ar

e ex

pre

ssed

as

mea

n ±

sta

nd

ard d

evia

tio

n (

n=

3);

valu

es

mar

ked

by t

he

dif

fere

nt

lett

er w

ithin

the

sam

e co

lum

n a

re

sig

nif

icantl

y d

iffe

rent

(P <

0.0

5).

TG

ase

trea

ted

suri

mi

gel

Bre

akin

g f

orc

e

(g)

Defo

rmat

ion

dis

tance

(m

m)

Coo

kin

g l

oss

(%)

colo

r

L

a b

Co

ntr

ol

438.5

±21.5

d

10.3

5±

0.4

7 b

5.8

5%

±0.3

5%

c 61.2

3±

1.3

3b

-0.3

3±

0.0

3d

7.1

9 ±

0.2

2d

0.1

U/g

593.4

±33.1

c 10.6

8±

0.4

8b

5.5

9%

±0.4

1%

c 62.8

5±

1.2

5b

-0.3

2±

0.0

2cd

8.1

6±

0.3

1c

0.5

U/g

683.2

±25.2

b

10.8

9±

0.5

1b

5.6

7%

±0.3

9%

c 61.3

3±

1.3

8b

-0.2

9±

0.0

2bc

7.9

5±

0.2

8c

1 U

/g

761.8

±21.2

a 11.3

3±

0.3

1a

5.8

9%

±0.4

5%

c 62.2

8±

2.3

1b

-0.3

7±

0.0

2e

8.3

9±

0.3

3bc

2 U

/g

728.5

±28.3

ab

8.3

6±

0.2

5c

10.5

5%

±0.9

8%

b

62.4

0±

2.1

4b

-0.2

7±

0.0

1b

8.8

1±

0.3

8b

4 U

/g

711.3

±27.6

b

8.0

7±

0.2

7c

13.6

9%

±1.8

6%

a 61.8

5±

1.8

3b

-0.1

5±

0.0

1a

8.7

9±

0.4

1b

Cat

fish

fil

let

785.4

±28.2

a 11.9

9±

0.3

2a

6.0

1±

0.4

2%

c 74.4

8±

1.2

3a

-0.3

9±

0.0

3d

13.6

4±

0.1

8a

177

5.4.7 SDS-PAGE of TGase-treated protein gel

Though the textural profiles of TGase treated protein gel provide the information

for evaluating the quality of protein gel, the protein profiles changes during TGase

treatment would provide the insight information to understand the protein chemistry of

TGase treated protein gel. The SDS-PAGE gel pf proteins extracted from catfish by-

products, and TGase treated myofibrillar proteins are shown in Figure 5.6. The results

showed that sarcoplasmic proteins were successfully separated from myofibrillar proteins

by pH 5.5 precipitation. The major bands of myofibrillar proteins are myosin heavy chain

and actin. The above results indicated that protein gel treated with 1 U/g of proteins had

the highest breaking force and deformation distances, and the breaking force and

deformation distance were decreased with TGase concentration go over 1 U/g of proteins.

Therefore, TGase of 1 U/g of proteins was chose for the following study. The TGase

treated myofibrillar proteins tends to form new large polymer (252.3KD) after 20 min

treatment at 40°C, the intensity of new formed polymer reaches a platform after 100 min

treatment. Several studies have shown that the breaking force and deformation distance

were increased as the addition of low concentration of TGase and decreased with the

usage of TGase above an optimum concentration (Ramırez, Rodrıguez-Sosa, Morales, &

Vázquez, 2000; Seighalani, Bakar, Saari, & Khoddami, 2017). TGase induces the

formation of non-disulfide bond, which results in the formation of myosin heavy chain

cross-linking and result to a stronger gel. We are the first to show the protein profiles

changes of protein gel made from catfish by-products during TGase treatment at different

time. In addition, most of the TGase treatment study had used the TGase based on the

weight of protein sol (Jiang, Hsieh, Ho, & Chung, 2000; Seighalani, Bakar, Saari, &

178

Khoddami, 2017). However, the protein content of protein sol may vary. It is more

reliable by using TGase based on the protein content in protein gel.

Figure 5.6 SDS-PAGE gel of proteins extracted from catfish by-products, and TGase-

treated protein gel.

Lane 1: Molecular marker; Lane 2: pH 8.5 extracted proteins; Lane 3: supernatant after

pH 5.5 precipitation; Lane 4: pH 5.5 precipitation; Lane 5: MTGase treatment control

group (pH 5.5 precipitation); Lane 6: TGase (1 U/g) treated for 10 min; Lane 7: TGase (1

U/g) treated for 20 min; Lane 8: TGase (1 U/g) treated for 40 min; Lane 9: TGase (1 U/g)

treated for 60 min; Lane 10: TGase (1 U/g) treated for 80 min; Lane 11: TGase (1 U/g)

treated for 100 min; Lane 12: TGase (1 U/g) treated for 120 min.

5.4.8 Scanning electron microscopy

SDS-PAGE gel provides the protein profiles of TGase treated protein gel.

However, it’s still hard to understand the gel structure of TGase treated protein gel, and

Myosin heavy chain

Actin

179

the differences between control (without TGase) and TGase treated protein gels.

Therefore, scanning electron microscopy was used to visualize the microstructure of

protein gel with and without TGase treatment. For comparison, commercial fish ball

(Wei-Chuan, USA) was obtained from local Asian market and used as positive control,

protein gel made from myofibrillar proteins (pH 8.5 extraction) without TGase treatment

was used as negative control. Protein gel treated with TGase at concentrations of 0, 0.1,

0.5, 1, 2, and 4 U/g of proteins for 2 hrs were used for imaging. The images obtained

from SEM are shown in Figure 5.7. The results indicated that as the increase of TGase

concentration from 0 to 1 U/g of proteins, the gel had finer, longer and denser strands

than those without transglutaminase. These strands could form the network with more

fibrillar structure. This observation could be evidenced by a higher breaking force (Table

5.1). And this observation was in agreement with previous study that TGase treated

protein gel from threadfin bream exhibited finer and denser gel network (Kaewudom,

Benjakul, & Kijroongrojana, 2013). Discontinuous network with large holes and grooves

was observed in control protein gel (without TGase) and a denser gel matrix has been

observed after addition of TGase (Chanarat & Benjakul, 2013; Hu, et al., 2015).

However, protein gels with the addition of 2 and 4 U/g of proteins were less compact

than that of 1 U/g of proteins, which indicated that excessive TGase could destroy the gel

network. Excessive transglutaminase would decrease the breaking force of surimi was

observed before (Seighalani, Bakar, Saari, & Khoddami, 2017). However, no

microstructure was reported in the above study which make comparison difficult. Pictures

of protein gels and commercial fish ball are shown in Figure 5.8 for reference.

180

Figure 5.7 Scanning electron microscopy images of TGase-treated protein gels.

A: commercial fish ball; B: protein gel without TGase; C: TGase treated protein gel (0.1

U/g); D: TGase treated protein gel (0.5 U/g); E: TGase treated protein gel (1 U/g); F:

TGase treated protein gel (2 U/g); G: TGase treated protein gel (4 U/g).

181

Figure 5.8 Pictures of commercial fish ball and protein gels made from catfish by-

products with (1 U/g) and without transglutaminase treatment.

Commercial fish ball (Wei-Chuan, USA) was obtained from a local Asian market.

5.5 Conclusion

Higher extraction pH could result in a higher protein extraction yield with the loss

of some integrity of proteins. And extraction pH had significant effect on the secondary

structure of extracted myofibrillar proteins. Myofibrillar proteins extracted with pH 8.5

gave the highest yield without compromise the integrity of extracted proteins.

Rheological profiles of protein sols made from myofibrillar proteins showed that protein

gel made from pH 8.5 extracted proteins was the best. TGase could improve the gel

strength, but excessive TGase would result a weaker gel matrix.

182

CHAPTER VI

SUMMARY OF ACHIEVEMENTS, LIMITATIONS AND FUTURE WORK

6.1 Summary

In summary, catfish by-products including catfish skins, heads and frames have

been successfully extracted into protein products and their properties were characterized.

Collagen was extracted from catfish skin and characterized, the extraction

condition has been optimized, and the extraction kinetics were investigated as well. The

extracted collagen could have applications in food, medical and cosmetic industries.

The mixture of catfish heads and frames was hydrolyzed by 8 proteases (plant or

bacteria derived) under different conditions. Ficin (80 AzU/g) was the most efficient in

hydrolyzing the ground catfish by-product (DH reaching 71.88%) in 120 min at 30°C

among all the enzymes. Thermolysin could be used for the industries to hydrolysis

protein by-products with the highest hydrolysis economy. Functional properties of

hydrolysates were affected by pH, and hydrolysates exhibited high emulsion properties

could be used as a replacement of SPI in food system.

The myofibrillar proteins were extracted from the mixture of catfish heads and

frames with different pH conditions, and made into protein gels. Transglutaminase

(TGase) was incorporated to improve the gel structure. The results indicated that

myofibrillar proteins could be extracted (pH 8.5 without compromise the integrity of

183

proteins) from catfish by-products and made into protein gels, which is a value-added

product.

6.2 Limitations and future work

The research flow diagrams in proposal are shown in Figure 3.7 (Chapter III) and

Figure 4.17 (Chapter IV). Simulated gastrointestinal digestion of protein hydrolysates

was not completed in the present study. The reason was that the proposal is very

ambitious and a new research section (Chapter V) was added to study the extract pH

effect on the myofibrillar proteins in by-products and their protein gels. Simulated

gastrointestinal digestion of the protein hydrolysates should be included in the future’s

study. Future work should be focused on the following areas:

(1) Overall: The data obtained in the present study were from laboratory

experiments. Scaling up experiments should be included in the further study.

(2) Overall: The residue after protein extraction or hydrolysis should be studied.

The residues might could be processed into calcium supplement.

(3) Chapter III: Porcine or bovine derived collagen should be included for

reference/comparison.

(4) Chapter IV: Simulated gastrointestinal digestion of the protein hydrolysates

should be studied in the future.

(5) Chapter IV: The amino acids sequence of the hydrolysates showed good

emulsion capacity, foaming capacity, ACE and DPP-IV inhibition activities should be

studied in the future.

184

(6) Chapter IV: Bitterness of the protein hydrolysates should be studied in the

future.

(7) Chapter V: The color of the protein gels made from myofibrillar proteins

extracted from catfish heads and frames was darker than the commercial surimi products

and should be improved in the future.

185

REFERENCES

Adhikari, B. B., Somerset, J., Stull, J. T., & Fajer, P. G. (1999). Dynamic modulation of

the regulatory domain of myosin heads by pH, ionic strength, and RLC

phosphorylation in synthetic myosin filaments. Biochemistry, 38, 3127-3132.

Adler-Nissen, J. (1977). Enzymatic hydrolysis of soy protein for nutritional fortification

of low pH food. In Annales de la nutrition et de l'alimentation, 32, 205-216.

Adler-Nissen, J. (1979). Determination of the degree of hydrolysis of food protein

hydrolysates by trinitrobenzenesulfonic acid. Journal of Agricultural and Food

Chemistry, 27, 1256.

Amiza, M., Nurul Ashikin, S., & Faazaz, A. (2011). Optimization of enzymatic protein

hydrolysis from silver catfish (Pangasius sp.) frame. International Food Research

Journal, 18, 775-781.

Arnesen, J. A., & Gildberg, A. (2007). Extraction and characterisation of gelatine from

Atlantic salmon (Salmo salar) skin. Bioresource Technology, 98, 53-57.

Asghar, A., & Pearson, A. (1980). Influence of ante-and postmortem treatments upon

muscle composition and meat quality. Advances in Food Research, 26, 53-213.

Avena-Bustillos, R., Olsen, C., Olson, D., Chiou, B., Yee, E., Bechtel, P., & McHugh, T.

(2006). Water vapor permeability of mammalian and fish gelatin films. Journal of

Food Science, 71, E202-E207.

Bastidas, O. H., Green, B., Sprague, M., & Peters, M. H. (2016). Few Ramachandran

Angle Changes Provide Interaction Strength Increase in Aβ42 versus Aβ40

Amyloid Fibrils. Scientific Reports, 6, 36499.

Beddows, C., & Ardeshir, A. (1979). The production of soluble fish protein solution for

use in fish sauce manufacture I. The use of added enzymes. International Journal

of Food Science & Technology, 14, 603-612.

Benhabiles, M., Abdi, N., Drouiche, N., Lounici, H., Pauss, A., Goosen, M., & Mameri,

N. (2012). Fish protein hydrolysate production from sardine solid waste by crude

pepsin enzymatic hydrolysis in a bioreactor coupled to an ultrafiltration unit.

Materials Science and Engineering: C, 32, 922-928.

186

Benjakul, S., & Morrissey, M. T. (1997). Protein hydrolysates from Pacific whiting solid

wastes. Journal of Agricultural and Food Chemistry, 45, 3423-3430.

Bergman, I., & Loxley, R. (1963). Two improved and simplified methods for the

spectrophotometric determination of hydroxyproline. Analytical Chemistry, 35,

1961-1965.

Bougatef, A., Nedjar-Arroume, N., Ravallec-Plé, R., Leroy, Y., Guillochon, D., Barkia,

A., & Nasri, M. (2008). Angiotensin I-converting enzyme (ACE) inhibitory

activities of sardinelle (Sardinella aurita) by-products protein hydrolysates

obtained by treatment with microbial and visceral fish serine proteases. Food

Chemistry, 111, 350-356.

Bradford, M. M. (1976). A rapid and sensitive method for the quantitation of microgram

quantities of protein utilizing the principle of protein-dye binding. Analytical

Biochemistry, 72, 248-254.

BSI (British Standards Institution) (1975). Methods for sampling and testing gelatin

(physical and chemical methods). London: BSI.

Cai, T., Chang, K.-C., & Lunde, H. (1996). Physicochemical properties and yields of

sunflower protein enzymatic hydrolysates as affected by enzyme and defatted

sunflower meal. Journal of Agricultural and Food Chemistry, 44, 3500-3506.

Cando, D., Herranz, B., Borderías, A. J., & Moreno, H. M. (2016). Different additives to

enhance the gelation of surimi gel with reduced sodium content. Food Chemistry,

196, 791-799.

Cao, Y., & Xiong, Y. L. (2015). Chlorogenic acid-mediated gel formation of oxidatively

stressed myofibrillar protein. Food Chemistry, 180, 235-243.

Chanarat, S., & Benjakul, S. (2013). Effect of formaldehyde on protein cross-linking and

gel forming ability of surimi from lizardfish induced by microbial

transglutaminase. Food Hydrocolloids, 30, 704-711.

Chen, J., Li, L., Yi, R., Xu, N., Gao, R., & Hong, B. (2016). Extraction and

characterization of acid-soluble collagen from scales and skin of tilapia

(Oreochromis niloticus). LWT - Food Science and Technology, 66, 453-459.

Chen, S., Chen, H., Xie, Q., Hong, B., Chen, J., Hua, F., Bai, K., He, J., Yi, R., & Wu, H.

(2016). Rapid isolation of high purity pepsin-soluble type I collagen from scales

of red drum fish (Sciaenops ocellatus). Food Hydrocolloids, 52, 468-477.

Chen, C. W., & Ho, C. T. (1995). Antioxidant properties of polyphenols extracted from

green and black teas. Journal of food lipids, 2, 35-46.

187

Cheow, C. S., Norizah, M. S., Kyaw, Z. Y., & Howell, N. K. (2007). Preparation and

characterisation of gelatins from the skins of sin croaker (Johnius dussumieri) and

shortfin scad (Decapterus macrosoma). Food Chemistry, 101, 386-391.

Cheung, I. W., & Li-Chan, E. C. (2017). Enzymatic production of protein hydrolysates

from steelhead (Oncorhynchus mykiss) skin gelatin as inhibitors of dipeptidyl-

peptidase IV and angiotensin-I converting enzyme. Journal of Functional Foods,

28, 254-264.

Choi, J. D., Kim, D. S., & Kim, J. S. (1998). Preparation of calcium-based powder from

fish bone and its characteristics. Applied Biological Chemistry, 41, 147-152.

Choi, S. S., & Regenstein, J. M. (2000). Physicochemical and Sensory Characteristics of

Fish Gelatin. Journal of Food Science, 65, 194-199.

Choi, Y., & Park, J. W. (2002). Acid-aided protein recovery from enzyme‐rich pacific

whiting. Journal of Food Science, 67, 2962-2967.

Coêlho, D. F., Saturnino, T. P., Fernandes, F. F., Mazzola, P. G., Silveira, E., &

Tambourgi, E. B. (2016). Azocasein Substrate for Determination of Proteolytic

Activity: Reexamining a Traditional Method Using Bromelain Samples. BioMed

Research International, 2016, 1-6.

Corzo-Martínez, M., Moreno, F. J., Villamiel, M., Patino, J. M. R., & Sánchez, C. C.

(2017). Effect of glycation and limited hydrolysis on interfacial and foaming

properties of bovine β-lactoglobulin. Food Hydrocolloids, 66, 16-26.

Davenport, M., & Kristinsson, H. (2011). Channel catfish (Ictalurus punctatus) muscle

protein isolate performance processed under different acid and alkali pH values.

Journal of Food Science, 76, E240-E247.

Deacon, C. F. (2018). Peptide degradation and the role of DPP-4 inhibitors in the

treatment of type 2 diabetes. Peptides, 100, 150-157.

Diniz, F., & Martin, A. (1997). Effects of the extent of enzymatic hydrolysis on

functional properties of shark protein hydrolysate. LWT-Food Science and

Technology, 30, 266-272.

Dong, S., Zeng, M., Wang, D., Liu, Z., Zhao, Y., & Yang, H. (2008). Antioxidant and

biochemical properties of protein hydrolysates prepared from Silver carp

(Hypophthalmichthys molitrix). Food chemistry, 107, 1485-1493.

Duan, R., Zhang, J., Du, X., Yao, X., & Konno, K. (2009). Properties of collagen from

skin, scale and bone of carp (Cyprinus carpio). Food Chemistry, 112, 702-706.

Eastoe, J. (1955). The amino acid composition of mammalian collagen and gelatin.

Biochemical Journal, 61, 589.

188

Egerton, S., Culloty, S., Whooley, J., Stanton, C., & Ross, R. P. (2018). Characterization

of protein hydrolysates from blue whiting (Micromesistius poutassou) and their

application in beverage fortification. Food Chemistry, 245, 698-706.

Engel, J. (1987). Folding and unfolding of collagen triple helices. Advances in Meat

Research, 4, 145-161.

Farvin, K. S., Andersen, L. L., Otte, J., Nielsen, H. H., Jessen, F., & Jacobsen, C. (2016).

Antioxidant activity of cod (Gadus morhua) protein hydrolysates: Fractionation

and characterisation of peptide fractions. Food Chemistry, 204, 409-419.

Foegeding, E., Lanier, T., & Hultin, H. (1996). Characteristics of edible muscle tissues.

Food Chemistry, 3, 879-942.

Gbogouri, G., Linder, M., Fanni, J., & Parmentier, M. (2004). Influence of hydrolysis

degree on the functional properties of salmon byproducts hydrolysates. Journal of

Food Science, 69, C615-C622.

Gehring, C., Gigliotti, J., Moritz, J., Tou, J., & Jaczynski, J. (2011). Functional and

nutritional characteristics of proteins and lipids recovered by isoelectric

processing of fish by-products and low-value fish: a review. Food Chemistry, 124,

422-431.

Gehring, C. K., Davenport, M. P., & Jaczynski, J. (2009). Functional and nutritional

quality of protein and lipid recovered from fish processing by-products and

underutilized aquatic species using isoelectric solubilization/precipitation.

Current Nutrition & Food Science, 5, 17-39.

Ghribi, A. M., Gafsi, I. M., Sila, A., Blecker, C., Danthine, S., Attia, H., Bougatef, A., &

Besbes, S. (2015). Effects of enzymatic hydrolysis on conformational and

functional properties of chickpea protein isolate. Food Chemistry, 187, 322-330.

Gildberg, A. (1992). Recovery of proteinases and protein hydrolysates from fish viscera.,

Bioresource Technology 39, 271-276.

Gilsenan, P., & Ross-Murphy, S. (2000). Rheological characterisation of gelatins from

mammalian and marine sources. Food Hydrocolloids, 14, 191-195.

Gómez-Guillén, M. C., Giménez, B., López-Caballero, M. E., & Montero, M. P. (2011).

Functional and bioactive properties of collagen and gelatin from alternative

sources: A review. Food Hydrocolloids, 25, 1813-1827.

Goodsell, D. S., Dutta, S., Zardecki, C., Voigt, M., Berman, H. M., & Burley, S. K.

(2015). The RCSB PDB “molecule of the month”: inspiring a molecular view of

biology. PLoS Biology, 13, e1002140.

189

Greenfield, N. J. (1999). Applications of circular dichroism in protein and peptide

analysis. TrAC Trends in Analytical Chemistry, 18, 236-244.

Greenfield, N. J. (2006). Using circular dichroism collected as a function of temperature

to determine the thermodynamics of protein unfolding and binding interactions.

Nature Protocols, 1, 2527.

Guerard, F., Dufosse, L., De La Broise, D., & Binet, A. (2001). Enzymatic hydrolysis of

proteins from yellowfin tuna (Thunnus albacares) wastes using Alcalase. Journal

of Molecular Catalysis B: Enzymatic, 11, 1051-1059.

Guo, M., Liu, S., Ismail, M., Farid, M.M., Ji, H., Mao, W., Gao, J. & Li, C. (2017).

Changes in the myosin secondary structure and shrimp surimi gel strength

induced by dense phase carbon dioxide. Food Chemistry, 227, 219-226.

Hall, F. G., Jones, O. G., O'Haire, M. E., & Liceaga, A. M. (2017). Functional properties

of tropical banded cricket (Gryllodes sigillatus) protein hydrolysates. Food

Chemistry, 224, 414-422.

Halldórsdóttir, S. M., Hamaguchi, P. Y., Sveinsdóttir, H., Kristinsson, H. G., Bergsson,

A. B., & Thorkelsson, G. (2011). Properties of hydrolysed saithe protein isolates

prepared via pH shift process with and without dewatering. LWT-Food Science

and Technology, 44, 1999-2004.

Halldorsdottir, S. M., Kristinsson, H. G., Sveinsdottir, H., Thorkelsson, G., &

Hamaguchi, P. Y. (2013). The effect of natural antioxidants on haemoglobin-

mediated lipid oxidation during enzymatic hydrolysis of cod protein. Food

Chemistry, 141, 914-919.

Halldorsdottir, S. M., Sveinsdottir, H., Gudmundsdottir, A., Thorkelsson, G., &

Kristinsson, H. G. (2014). High quality fish protein hydrolysates prepared from

by-product material with Fucus vesiculosus extract. Journal of Functional

Foods, 9, 10-17.

Hathwar, S. C., Bijinu, B., Rai, A. K., & Narayan, B. (2011). Simultaneous recovery of

lipids and proteins by enzymatic hydrolysis of fish industry waste using different

commercial proteases. Applied Biochemistry and Biotechnology, 164, 115-124.

Hoyle., & Merritt, J. (1994). Quality of fish protein hydrolysates from herring (Clupea

harengus). Journal of Food Science, 59, 76-79.

Hsu, K.-C., Li-Chan, E. C., & Jao, C.-L. (2011). Antiproliferative activity of peptides

prepared from enzymatic hydrolysates of tuna dark muscle on human breast

cancer cell line MCF-7. Food Chemistry, 126, 617-622.

190

Hu, Y., Liu, W., Yuan, C., Morioka, K., Chen, S., Liu, D., & Ye, X. (2015).

Enhancement of the gelation properties of hairtail (Trichiurus haumela) muscle

protein with curdlan and transglutaminase. Food Chemistry, 176, 115-122.

Hu, Y., Shao, Y., Wu, C., Yuan, C., Ishimura, G., Liu, W., & Chen, S. (2018). γ-PGA

and MTGase improve the formation of ε-(γ-glutamyl) lysine cross-links within

hairtail (Trichiurus haumela) surimi protein. Food Chemistry, 242, 330-337.

Huang, C.-Y., Kuo, J.-M., Wu, S.-J., & Tsai, H.-T. (2016). Isolation and characterization

of fish scale collagen from tilapia (Oreochromis sp.) by a novel extrusion–hydro-

extraction process. Food Chemistry, 190, 997-1006.

Huang, Y.-w., Abdel-Aal, H. A., & Awad, A. H. (1996). Properties of surimi made from

tilapia. In Nineteenth and twentieth annual conferences of Tropical and

Subtropical Seafood Science and Technology Society of the Americans).

Ikoma, T., Kobayashi, H., Tanaka, J., Walsh, D., & Mann, S. (2003). Physical properties

of type I collagen extracted from fish scales of Pagrus major and Oreochromis

niloticas. International Journal of Biological Macromolecules, 32, 199-204.

Imm, J., & Lee, C. (1999). Production of seafood flavor from red hake (Urophycis chuss)

by enzymatic hydrolysis. Journal of Agricultural and Food Chemistry, 47, 2360-

2366.

Iwasaki, T., Noshiroya, K., Saitoh, N., Okano, K., & Yamamoto, K. (2006). Studies of

the effect of hydrostatic pressure pretreatment on thermal gelation of chicken

myofibrils and pork meat patty. Food Chemistry, 95, 474-483.

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

(2010). Influence of degree of hydrolysis on functional properties, antioxidant

activity and ACE inhibitory activity of peanut protein hydrolysate. Food

Chemistry, 121, 178-184.

Jamilah, B., & Harvinder, K. G. (2002). Properties of gelatins from skins of fish—black

tilapia (Oreochromis mossambicus) and red tilapia (Oreochromis nilotica). Food

Chemistry, 77, 81-84.

Jenkelunas, P. J., & Li-Chan, E. C. (2018). Production and assessment of Pacific hake

(Merluccius productus) hydrolysates as cryoprotectants for frozen fish mince.

Food Chemistry, 239, 535-543.

Jia, G., Nirasawa, S., Ji, X., Luo, Y., & Liu, H. (2018). Physicochemical changes in

myofibrillar proteins extracted from pork tenderloin thawed by a high-voltage

electrostatic field. Food Chemistry, 240, 910-916.

191

Jiang, S. T., Hsieh, J. F., Ho, M. L., & Chung, Y. C. (2000). Microbial Transglutaminase

Affects Gel Properties of Golden Threadfin‐bream and Pollack Surimi. Journal of

Food Science, 65, 694-699.

Johnston-Bank, F. (1983). From tannery to table: an account of gelatin production.

Journal of the Society of Leather Technologists and Chemists, 68, 141-145.

Jongjareonrak, A., Benjakul, S., Visessanguan, W., Nagai, T., & Tanaka, M. (2005).

Isolation and characterisation of acid and pepsin-solubilised collagens from the

skin of Brownstripe red snapper (Lutjanus vitta). Food Chemistry, 93, 475-484.

Kaewudom, P., Benjakul, S., & Kijroongrojana, K. (2013). Properties of surimi gel as

influenced by fish gelatin and microbial transglutaminase. Food Bioscience, 1,

39-47.

Karim, A. A., & Bhat, R. (2009). Fish gelatin: properties, challenges, and prospects as an

alternative to mammalian gelatins. Food Hydrocolloids, 23, 563-576.

Kimura, S., ZHU, X. P., Matsui, R., Shijoh, M., & Takamizawa, S. (1988).

Characterization of fish muscle type I collagen. Journal of Food Science, 53,

1315-1318.

Kittiphattanabawon, P., Benjakul, S., Visessanguan, W., Kishimura, H., & Shahidi, F.

(2010). Isolation and characterisation of collagen from the skin of brownbanded

bamboo shark (Chiloscyllium punctatum). Food Chemistry, 119, 1519-1526.

Kittiphattanabawon, P., Benjakul, S., Visessanguan, W., Nagai, T., & Tanaka, M. (2005).

Characterisation of acid-soluble collagen from skin and bone of bigeye snapper

(Priacanthus tayenus). Food Chemistry, 89, 363-372.

Klompong, V., Benjakul, S., Kantachote, D., & Shahidi, F. (2007). Antioxidative activity

and functional properties of protein hydrolysate of yellow stripe trevally

(Selaroides leptolepis) as influenced by the degree of hydrolysis and enzyme type.

Food Chemistry, 102, 1317-1327.

Kong, F., Chang, S., Liu, Z., & Wilson, L. (2008). Changes of soybean quality during

storage as related to soymilk and tofu making. Journal of Food Science, 73, S134-

S144.

Kok, N., Thawornchinsombut, S., Park, J. W. (2013). Manufacture of fish ball. Surimi

and Surimi Seafood (Third Edition), Chapter 11, 291.

Kristinsson, H. G., Lanier, T. C., Halldorsdottir, S. M., Geirsdottir, M., & Park, J. W.

(2013). Fish protein isolate by pH shift. Surimi and surimi seafood, 169-192.

192

Kristinsson, H. G., & Liang, Y. (2006). Effect of pH‐Shift Processing and Surimi

Processing on Atlantic Croaker (Micropogonias undulates) Muscle Proteins.

Journal of Food Science, 71, C304-C312.

Kristinsson, H. G., & Rasco, B. A. (2000a). Fish protein hydrolysates: production,

biochemical, and functional properties. Critical Reviews in Food Science and

Nutrition, 40, 43-81.

Kristinsson, H. G., & Rasco, B. A. (2000b). Kinetics of the hydrolysis of Atlantic salmon

(Salmo salar) muscle proteins by alkaline proteases and a visceral serine protease

mixture. Process Biochemistry, 36, 131-139.

Kristinsson, H. G., Theodore, A. E., Demir, N., & Ingadottir, B. (2005). A comparative

study between acid‐and alkali‐aided processing and surimi processing for the

recovery of proteins from Channel catfish muscle. Journal of Food Science, 70,

C298-C306.

Kubota, S., Tamura, Y., Morioka, K., & Itoh, Y. (2003). Variable Pressure‐scanning

Electron Microscopic Observation of Walleye Pollack Surimi Gel. Journal of

Food Science, 68, 307-311.

Kumar, D., & BHALLA, T. C. (2003). Bacillus sp. APR-4 protease in protein recovery

from waste bones. National Academy Science Letters, 26, 332-335.

Kuraishi, C., Sakamoto, J., & Soeda, T. (1996). The usefulness of transglutaminase for

food processing. In): ACS Publications, Chapter 3, 29-38.

Ladrat, C., Verrez-Bagnis, V., Noël, J., & Fleurence, J. (2003). In vitro proteolysis of

myofibrillar and sarcoplasmic proteins of white muscle of sea bass

(Dicentrarchus labrax L.): effects of cathepsins B, D and L. Food Chemistry, 81,

517-525.

Laemmli, U. K. (1970). Cleavage of structural proteins during the assembly of the head

of bacteriophage T4. Nature, 227, 680-685.

Lansdowne, L., Beamer, S., Jaczynski, J., & Matak, K. (2009). Survival of Listeria

innocua after isoelectric solubilization and precipitation of fish protein. Journal of

Food Science, 74, M201-M205.

Le Maux, S., Nongonierma, A. B., Murray, B., Kelly, P. M., & FitzGerald, R. J. (2015).

Identification of short peptide sequences in the nanofiltration permeate of a

bioactive whey protein hydrolysate. Food Research International, 77, 534-539.

Liaset, B., Lied, E., & Espe, M. (2000). Enzymatic hydrolysis of by‐products from the

fish‐filleting industry; chemical characterisation and nutritional evaluation.

Journal of the Science of Food and Agriculture, 80, 581-589.

193

Liceaga‐Gesualdo, A., & Li‐Chan, E. (1999). Functional properties of fish protein

hydrolysate from herring (Clupea harengus). Journal of Food Science, 64, 1000-

1004.

Li-Chan, E. C., Hunag, S. L., Jao, C. L., Ho, K. P., & Hsu, K. C. (2012). Peptides derived

from Atlantic salmon skin gelatin as dipeptidyl-peptidase IV inhibitors. Journal of

Agricultural and Food Chemistry, 60, 973-978.

Liu, D., Wei, G., Li, T., Hu, J., Lu, N., Regenstein, J. M., & Zhou, P. (2015). Effects of

alkaline pretreatments and acid extraction conditions on the acid-soluble collagen

from grass carp (Ctenopharyngodon idella) skin. Food Chemistry, 172, 836-843.

Liu, H., Li, D., & Guo, S. (2007). Studies on collagen from the skin of Channel catfish

(Ictalurus punctaus). Food Chemistry, 101, 621-625.

Liu, R., Zhao, S., Xiong, S., Xie, B., & Liu, H. (2007). Studies on fish and pork paste

gelation by dynamic rheology and circular dichroism. Journal of Food Science,

72, E399-E403.

Liu, W., Li, G., Miao, Y., & Wu, X. (2009). Preparation and characterization of pepsin-

solubilized type I collagen from the scales of snakehead (Ophiocephalus Argus).

Journal of Food Biochemistry, 33, 20-37.

Liu, Y., Li, X., Chen, Z., Yu, J., Wang, F., & Wang, J. (2014). Characterization of

structural and functional properties of fish protein hydrolysates from surimi

processing by-products. Food Chemistry, 151, 459-465.

Ma, X.-S., Yi, S.-M., Yu, Y.-M., Li, J.-R., & Chen, J.-R. (2015). Changes in gel

properties and water properties of Nemipterus virgatus surimi gel induced by

high-pressure processing. LWT-Food Science and Technology, 61, 377-384.

Mackie, I. M. (1974). Proteolytic enzymes in recovery of proteins from fish waste.

Process Biochemistry (UK).

McAlpin, C. R., Dillard, J., Kim, J., & Montanez, J. (1994). An economic analysis of

producing surimi from catfish byproducts. Bulletin (Mississippi Agricultural and

Forestry Experiment Station)(USA).

Mendis, E., Rajapakse, N., & Kim, S.-K. (2005). Antioxidant properties of a radical-

scavenging peptide purified from enzymatically prepared fish skin gelatin

hydrolysate. Journal of Agricultural and Food Chemistry, 53, 581-587.

Montero, P., Alvarez, C., Marti, M., & Borderias, A. (1995). Plaice skin collagen

extraction and functional properties. Journal of Food Science, 60, 1-3.

194

Montero, P., Gómez-Guillén, M., & Borderıas, A. (1999). Functional characterisation of

muscle and skin collagenous material from hake (Merluccius merluccius L.).

Food Chemistry, 65, 55-59.

Montero, P., & Gómez-Guillén, M. C. (2000). Extracting Conditions for Megrim

(Lepidorhombus boscii) Skin Collagen Affect Functional Properties of the

Resulting Gelatin. Journal of Food Science, 65, 434-438.

Moon, J. H., Yoon, W. B., & Park, J. W. (2017). Assessing the textural properties of

Pacific whiting and Alaska pollock surimi gels prepared with carrot under various

heating rates. Food Bioscience, 20, 12-18.

Morales-Medina, R., Tamm, F., Guadix, A., Guadix, E., & Drusch, S. (2016). Functional

and antioxidant properties of hydrolysates of sardine (S. pilchardus) and horse

mackerel (T. mediterraneus) for the microencapsulation of fish oil by spray-

drying. Food Chemistry, 194, 1208-1216.

Murthy, P. S., Rai, A. K., & Bhaskar, N. (2014). Fermentative recovery of lipids and

proteins from freshwater fish head waste with reference to antimicrobial and

antioxidant properties of protein hydrolysate. Journal of Food Science and

Technology, 51, 1884-1892.

Mutilangi, W., Panyam, D., & Kilara, A. (1996). Functional properties of hydrolysates

from proteolysis of heat‐denatured whey protein isolate. Journal of Food Science,

61, 270-275.

Muyonga, J., Cole, C., & Duodu, K. (2004a). Characterisation of acid soluble collagen

from skins of young and adult Nile perch (Lates niloticus). Food Chemistry, 85,

81-89.

Muyonga, J., Cole, C., & Duodu, K. (2004b). Fourier transform infrared (FTIR)

spectroscopic study of acid soluble collagen and gelatin from skins and bones of

young and adult Nile perch (Lates niloticus). Food Chemistry, 86, 325-332.

Muyonga, J. H., Cole, C. G. B., & Duodu, K. G. (2004a). Characterisation of acid soluble

collagen from skins of young and adult Nile perch (Lates niloticus). Food

Chemistry, 85, 81-89.

Muyonga, J. H., Cole, C. G. B., & Duodu, K. G. (2004b). Extraction and physico-

chemical characterisation of Nile perch (Lates niloticus) skin and bone gelatin.

Food Hydrocolloids, 18, 581-592.

Nagai, T., Araki, Y., & Suzuki, N. (2002). Collagen of the skin of ocellate puffer fish

(Takifugu rubripes). Food Chemistry, 78, 173-177.

Nagai, T., & Suzuki, N. (2000). Isolation of collagen from fish waste material — skin,

bone and fins. Food Chemistry, 68, 277-281.

195

Nagai, T., & Suzuki, N. (2002). Preparation and partial characterization of collagen from

paper nautilus (Argonauta argo, Linnaeus) outer skin. Food Chemistry, 76, 149-

153.

Nagai, T., Yamashita, E., Taniguchi, K., Kanamori, N., & Suzuki, N. (2001). Isolation

and characterisation of collagen from the outer skin waste material of cuttlefish

(Sepia lycidas). Food Chemistry, 72, 425-429.

Nakajima, K., Yoshie-Stark, Y., & Ogushi, M. (2009). Comparison of ACE inhibitory

and DPPH radical scavenging activities of fish muscle hydrolysates. Food

Chemistry, 114, 844-851.

Nalinanon, S., Benjakul, S., Visessanguan, W., & Kishimura, H. (2007). Use of pepsin

for collagen extraction from the skin of bigeye snapper (Priacanthus tayenus).

Food Chemistry, 104, 593-601.

National Fishery Institute. (2017). Top ten most popular seafood items.

https://www.aboutseafood.com/press_release/top-10-list-highlights-seafood-

consumption-progress/.

Ndez-Diaaz, C. F., & Fea, M. (1999). Optimization of the protein fraction of

microcapsules used in feeding of marine fish larvae using in vitro digestibility

techniques. Aquaculture Nutrition, 5, 107-113.

Nilsang, S., Lertsiri, S., Suphantharika, M., & Assavanig, A. (2005). Optimization of

enzymatic hydrolysis of fish soluble concentrate by commercial proteases.

Journal of Food Engineering, 70, 571-578.

Nolsøe, H., & Undeland, I. (2009). The acid and alkaline solubilization process for the

isolation of muscle proteins: state of the art. Food and Bioprocess Technology, 2,

1-27.

Nongonierma, A. B., & FitzGerald, R. J. (2013). Dipeptidyl peptidase IV inhibitory and

antioxidative properties of milk protein-derived dipeptides and

hydrolysates. Peptides, 39, 157-163.

Nongonierma, A. B., Paolella, S., Mudgil, P., Maqsood, S., & FitzGerald, R. J. (2017).

Dipeptidyl peptidase IV (DPP-IV) inhibitory properties of camel milk protein

hydrolysates generated with trypsin. Journal of Functional Foods, 34, 49-58.

Ogawa, M., Moody, M. W., Portier, R. J., Bell, J., Schexnayder, M. A., & Losso, J. N.

(2003). Biochemical properties of black drum and sheepshead seabream skin

collagen. Journal of Agricultural and Food Chemistry, 51, 8088-8092.

196

Ogawa, M., Portier, R. J., Moody, M. W., Bell, J., Schexnayder, M. A., & Losso, J. N.

(2004). Biochemical properties of bone and scale collagens isolated from the

subtropical fish black drum (Pogonia cromis) and sheepshead seabream

(Archosargus probatocephalus). Food Chemistry, 88, 495-501.

Ojha, K. S., Alvarez, C., Kumar, P., O'Donnell, C. P., & Tiwari, B. K. (2016). Effect of

enzymatic hydrolysis on the production of free amino acids from boarfish

(Capros aper) using second order polynomial regression models. LWT-Food

Science and Technology, 68, 470-476.

Onoue, Y., & Riddle, V. (1973). Use of plastein reaction in recovering protein from fish

waste. Journal of the Fisheries Board of Canada, 30, 1745-1747.

Pal, G. K., Nidheesh, T., & Suresh, P. V. (2015). Comparative study on characteristics

and in vitro fibril formation ability of acid and pepsin soluble collagen from the

skin of catla (Catla catla) and rohu (Labeo rohita). Food Research International,

76, 804-812.

Park, J. W. (2013). Surimi and surimi seafood: CRC press.

Park, J. W., Nozaki, H., Suzuki, T., & Beliveau, J. (2013). Historical review of Surimi

technology and market developments. Surimi and Surimi Seafood, 1, 1.

Park, S. H., CHO, S. Y., Kimura, M., Nozawa, H., & Seki, N. (2005). Effects of

microbial transglutaminase and starch on the thermal gelation of salted squid

muscle paste. Fisheries Science, 71, 896-903.

Pati, F., Adhikari, B., & Dhara, S. (2010). Isolation and characterization of fish scale

collagen of higher thermal stability. Bioresource Technology, 101, 3737-3742.

Pearce, K. N., & Kinsella, J. E. (1978). Emulsifying properties of proteins: evaluation of

a turbidimetric technique. Journal of Agricultural and Food Chemistry, 26, 716-

723.

Peleg, M. (1988). An empirical model for the description of moisture sorption curves.

Journal of Food Science, 53, 1216-1217.

Pietrasik, Z. (2003). Binding and textural properties of beef gels processed with κ-

carrageenan, egg albumin and microbial transglutaminase. Meat Science, 63, 317-

324.

Piez, K. A., & Gross, J. (1960). The amino acid composition of some fish collagens: The

relation between composition and structure. Journal of Biological Chemistry, 235,

995-998.

197

Prior, R. L., Wu, X., & Schaich, K. (2005). Standardized methods for the determination

of antioxidant capacity and phenolics in foods and dietary supplements. Journal

of Agricultural and Food Chemistry, 53, 4290-4302.

Qin, L., Zhu, B.W., Zhou, D.Y., Wu, H.T., Tan, H., Yang, J.F., Li, D.M., Dong, X.P. and

Murata, Y. (2011). Preparation and antioxidant activity of enzymatic hydrolysates

from purple sea urchin (Strongylocentrotus nudus) gonad. LWT-Food Science and

Technology, 44, 1113-1118.

Quaglia, G., & Orban, E. (1990). Influence of enzymatic hydrolysis on structure and

emulsifying properties of sardine (Sardina pilchardus) protein hydrolysates.

Journal of Food Science, 55, 1571-1573.

Raghavan, S., & Kristinsson, H. G. (2008). Conformational and rheological changes in

catfish myosin during alkali-induced unfolding and refolding. Food Chemistry,

107, 385-398.

Raghavan, S., & Kristinsson, H. G. (2009). ACE-inhibitory activity of tilapia protein

hydrolysates. Food Chemistry, 117, 582-588.

Rahali, V., Chobert, J. M., Haertle, T., & Gueguen, J. (2000). Emulsification of chemical

and enzymatic hydrolysates of β‐lactoglobulin: characterization of the peptides

adsorbed at the interface. Molecular Nutrition & Food Research, 44, 89-95.

Rai, A. K., Swapna, H. C., Bhaskar, N., Halami, P. M., & Sachindra, N. M. (2010).

Effect of fermentation ensilaging on recovery of oil from fresh water fish

viscera. Enzyme and Microbial Technology, 46, 9-13.

Ramırez, J. A., Rodrıguez-Sosa, R., Morales, O. G., & Vázquez, M. (2000). Surimi gels

from striped mullet (Mugil cephalus) employing microbial transglutaminase.

Food Chemistry, 70, 443-449.

Rebeca, B. D., Pena‐Vera, M., & Diaz‐Castaneda, M. (1991). Production of fish protein

hydrolysates with bacterial proteases; yield and nutritional value. Journal of Food

Science, 56, 309-314.

Rojas, M. J., Siqueira, P. F., Miranda, L. C., Tardioli, P. W., & Giordano, R. L. (2014).

Sequential proteolysis and cellulolytic hydrolysis of soybean hulls for

oligopeptides and ethanol production. Industrial Crops and Products, 61, 202-

210.

Salampessy, J., Phillips, M., Seneweera, S., & Kailasapathy, K. (2010). Release of

antimicrobial peptides through bromelain hydrolysis of leatherjacket (Meuchenia

sp.) insoluble proteins. Food Chemistry, 120, 556-560.

198

Salampessy, J., Reddy, N., Phillips, M., & Kailasapathy, K. (2017). Isolation and

characterization of nutraceutically potential ACE-Inhibitory peptides from

leatherjacket (Meuchenia sp.) protein hydrolysates. LWT-Food Science and

Technology, 80, 430-436.

Sangsawad, P., Roytrakul, S., Choowongkomon, K., Kitts, D.D., Chen, X.M., Meng, G.,

Li-Chan, E.C. and Yongsawatdigul, J. (2018). Transepithelial transport across

Caco-2 cell monolayers of angiotensin converting enzyme (ACE) inhibitory

peptides derived from simulated in vitro gastrointestinal digestion of cooked

chicken muscles. Food Chemistry, 251, 77-85.

Sathe, S., & Salunkhe, D. (1981). Functional properties of the great northern bean

(Phaseolus vulgaris L.) proteins: emulsion, foaming, viscosity, and gelation

properties. Journal of Food Science, 46, 71-81.

Sato, K. (1993). Comparative biochemistry of molecular species of fish and mammalian

collagens. Trends Comp. Biochem. Physiol, 1, 557-567.

Sato, K., Ebihara, T., Adachi, E., Kawashima, S., Hattori, S., & Irie, S. (2000). Possible

involvement of aminotelopeptide in self-assembly and thermal stability of

collagen I as revealed by its removal with proteases. Journal of Biological

Chemistry, 275, 25870-25875.

Seighalani, F. Z. B., Bakar, J., Saari, N., & Khoddami, A. (2017). Thermal and

physicochemical properties of red tilapia (Oreochromis niloticus) surimi gel as

affected by microbial transglutaminase. Animal Production Science, 57, 993-

1000.

Sen, D., Sripathy, N., Lahiry, N., Sreenivasan, A., & Subrahmanyan, V. (1962). Fish

hydrolysates. I. Rate of hydrolysis of fish flesh with papain. Food Technology, 16,

138-141.

Shahidi, F., Han, X.-Q., & Synowiecki, J. (1995). Production and characteristics of

protein hydrolysates from capelin (Mallotus villosus). Food Chemistry, 53, 285-

293.

Sila, A., Alvarez, O. M., Haddar, A., Frikha, F., Dhulster, P., Nedjar-Arroume, N., &

Bougatef, A. (2016). Purification, identification and structural modelling of DPP-

IV inhibiting peptides from barbel protein hydrolysate. Journal of

Chromatography B, 1008, 260-269.

Singh, P., Benjakul, S., Maqsood, S., & Kishimura, H. (2011). Isolation and

characterisation of collagen extracted from the skin of striped catfish

(Pangasianodon hypophthalmus). Food Chemistry, 124, 97-105.

199

Skierka, E., & Sadowska, M. (2007). The influence of different acids and pepsin on the

extractability of collagen from the skin of Baltic cod (Gadus morhua). Food

Chemistry, 105, 1302-1306.

Sodini, I., Lucas, A., Tissier, J., & Corrieu, G. (2005). Physical properties and

microstructure of yoghurts supplemented with milk protein hydrolysates.

International Dairy Journal, 15, 29-35.

Sripathy, N., Kadkol, S., Sen, D., Swaminathan, M., & Lahiry, N. (1963). Fish

hydrolysates. III. Influence of degree of hydrolysis on nutritive value. Journal of

Food Science, 28, 365-369.

Sripathy, N., Sen, D., Lahiry, N. L., Sreenivasan, A., & Subrahmanyan, V. (1962). Fish

hydrolysates. II. Standardisation of digestion conditions for preparation of

hydrolysates rich in peptones and proteoses. Food Technology, 16, 141-142.

Tadpitchayangkoon, P., Park, J. W., & Yongsawatdigul, J. (2010). Conformational

changes and dynamic rheological properties of fish sarcoplasmic proteins treated

at various pHs. Food Chemistry, 121, 1046-1052.

Tan, Y., Chang, S. K., & Zhang, Y. (2017). Comparison of α-amylase, α-glucosidase and

lipase inhibitory activity of the phenolic substances in two black legumes of

different genera. Food Chemistry, 214, 259-268.

Tan, Y., & Chang, S. K. (2018). Isolation and characterization of collagen extracted from

Channel catfish (Ictalurus punctatus) skin. Food Chemistry, 242, 147-155.

Tang, L., Chen, S., Su, W., Weng, W., Osako, K., & Tanaka, M. (2015). Physicochemical

properties and film-forming ability of fish skin collagen extracted from different

freshwater species. Process Biochemistry, 50, 148-155.

Theodore, A. E., & Kristinsson, H. G. (2007). Angiotensin converting enzyme inhibition

of fish protein hydrolysates prepared from alkaline‐aided Channel catfish protein

isolate. Journal of the Science of Food and Agriculture, 87, 2353-2357.

Tsai, J. S., Chen, J. L., & Pan, B. S. (2008). ACE-inhibitory peptides identified from the

muscle protein hydrolysate of hard clam (Meretrix lusoria). Process

Biochemistry, 43, 743-747.

Tulipano, G., Sibilia, V., Caroli, A. M., & Cocchi, D. (2011). Whey proteins as source of

dipeptidyl dipeptidase IV (dipeptidyl peptidase-4) inhibitors. Peptides, 32, 835-

838.

Þorkelsson, G., & Kristinsson, H. G. (2009). Bioactive peptides from marine sources.

State of art. Report to the NORA fund.

200

Tilman, D., Balzer, C., Hill, J., & Befort, B. L. (2011). Global food demand and the

sustainable intensification of agriculture. Proceedings of the National Academy of

Sciences, 108, 20260-20264.

Uchida, M., Ohshiba, Y., & Mogami, O. (2011). Novel dipeptidyl peptidase-4–inhibiting

peptide derived from β-lactoglobulin. Journal of Pharmacological Sciences, 117,

63-66.

United States Department of Agriculture/National Agricultural Statistics Service

(USDA/NASS). (1997). Production and Inventory of Channel Catfish. United

States Department of Agriculture, Washington, DC,

http://usda.mannlib.cornell.edu/usda/nass/CatfProd//1990s/1997/CatfProd-02-03-

1997.pdf.

United States Department of Agriculture/National Agricultural Statistics Service

(USDA/NASS). (1998). Production and Inventory of Channel Catfish. United

States Department of Agriculture, Washington, DC,

http://usda.mannlib.cornell.edu/usda/nass/CatfProd//1990s/1998/CatfProd-04-29-

1998.pdf.

United States Department of Agriculture/National Agricultural Statistics Service

(USDA/NASS). (1999). Production and Inventory of Channel Catfish. United

States Department of Agriculture, Washington, DC,

http://usda.mannlib.cornell.edu/usda/nass/CatfProd//1990s/1999/CatfProd-02-03-

1999.pdf.

United States Department of Agriculture/National Agricultural Statistics Service

(USDA/NASS). (2000). Production and Inventory of Channel Catfish. United

States Department of Agriculture, Washington, DC,

http://usda.mannlib.cornell.edu/usda/nass/CatfProd//2000s/2000/CatfProd-02-14-

2000.pdf.

United States Department of Agriculture/National Agricultural Statistics Service

(USDA/NASS). (2001). Production and Inventory of Channel Catfish. United

States Department of Agriculture, Washington, DC,

http://usda.mannlib.cornell.edu/usda/nass/CatfProd//2000s/2001/CatfProd-02-28-

2001.pdf.

United States Department of Agriculture/National Agricultural Statistics Service

(USDA/NASS). (2002). Production and Inventory of Channel Catfish. United

States Department of Agriculture, Washington, DC,

http://usda.mannlib.cornell.edu/usda/nass/CatfProd//2000s/2002/CatfProd-02-07-

2002.pdf.

201

United States Department of Agriculture/National Agricultural Statistics Service

(USDA/NASS). (2003). Production and Inventory of Channel Catfish. United

States Department of Agriculture, Washington, DC,

http://usda.mannlib.cornell.edu/usda/nass/CatfProd//2000s/2003/CatfProd-02-06-

2003.pdf.

United States Department of Agriculture/National Agricultural Statistics Service

(USDA/NASS). (2004). Production and Inventory of Channel Catfish. United

States Department of Agriculture, Washington, DC,

http://usda.mannlib.cornell.edu/usda/nass/CatfProd//2000s/2004/CatfProd-02-05-

2004.pdf.

United States Department of Agriculture/National Agricultural Statistics Service

(USDA/NASS). (2005). Production and Inventory of Channel Catfish. United

States Department of Agriculture, Washington, DC,

http://usda.mannlib.cornell.edu/usda/nass/CatfProd//2000s/2005/CatfProd-02-03-

2005.pdf.

United States Department of Agriculture/National Agricultural Statistics Service

(USDA/NASS). (2006). Production and Inventory of Channel Catfish. United

States Department of Agriculture, Washington, DC,

http://usda.mannlib.cornell.edu/usda/nass/CatfProd//2000s/2006/CatfProd-02-01-

2006.pdf.

United States Department of Agriculture/National Agricultural Statistics Service

(USDA/NASS). (2007). Production and Inventory of Channel Catfish. United

States Department of Agriculture, Washington, DC,

http://usda.mannlib.cornell.edu/usda/nass/CatfProd//2000s/2007/CatfProd-01-31-

2007.pdf.

United States Department of Agriculture/National Agricultural Statistics Service

(USDA/NASS). (2008). Production and Inventory of Channel Catfish. United

States Department of Agriculture, Washington, DC,

http://usda.mannlib.cornell.edu/usda/nass/CatfProd//2000s/2008/CatfProd-01-30-

2008_revision.pdf.

United States Department of Agriculture/National Agricultural Statistics Service

(USDA/NASS). (2009). Production and Inventory of Channel Catfish. United

States Department of Agriculture, Washington, DC,

http://usda.mannlib.cornell.edu/usda/nass/CatfProd//2000s/2009/CatfProd-01-30-

2009_revision.pdf.

202

United States Department of Agriculture/National Agricultural Statistics Service

(USDA/NASS). (2010). Production and Inventory of Channel Catfish. United

States Department of Agriculture, Washington, DC,

http://usda.mannlib.cornell.edu/usda/nass/CatfProd//2010s/2010/CatfProd-01-29-

2010.pdf.

United States Department of Agriculture/National Agricultural Statistics Service

(USDA/NASS). (2011). Production and Inventory of Channel Catfish. United

States Department of Agriculture, Washington, DC,

http://usda.mannlib.cornell.edu/usda/nass/CatfProd//2010s/2011/CatfProd-01-28-

2011.pdf.

United States Department of Agriculture/National Agricultural Statistics Service

(USDA/NASS). (2012). Production and Inventory of Channel Catfish. United

States Department of Agriculture, Washington, DC,

http://usda.mannlib.cornell.edu/usda/nass/CatfProd//2010s/2012/CatfProd-01-27-

2012.pdf.

United States Department of Agriculture/National Agricultural Statistics Service

(USDA/NASS). (2013). Production and Inventory of Channel Catfish. United

States Department of Agriculture, Washington, DC,

http://usda.mannlib.cornell.edu/usda/nass/CatfProd//2010s/2013/CatfProd-01-28-

2013.pdf.

United States Department of Agriculture/National Agricultural Statistics Service

(USDA/NASS). (2014). Production and Inventory of Channel Catfish. United

States Department of Agriculture, Washington, DC,

http://usda.mannlib.cornell.edu/usda/nass/CatfProd//2010s/2014/CatfProd-02-20-

2014.pdf.

United States Department of Agriculture/National Agricultural Statistics Service

(USDA/NASS). (2015). Production and Inventory of Channel Catfish. United

States Department of Agriculture, Washington, DC,

http://usda.mannlib.cornell.edu/usda/nass/CatfProd//2010s/2015/CatfProd-01-30-

2015.pdf.

United States Department of Agriculture/National Agricultural Statistics Service

(USDA/NASS). (2016). Production and Inventory of Channel Catfish. United

States Department of Agriculture, Washington, DC,

http://usda.mannlib.cornell.edu/usda/nass/CatfProd//2010s/2016/CatfProd-02-05-

2016.pdf.

203

United States Department of Agriculture/National Agricultural Statistics Service

(USDA/NASS). (2017). Production and Inventory of Channel Catfish. United

States Department of Agriculture, Washington, DC,

http://usda.mannlib.cornell.edu/usda/nass/CatfProd//2010s/2017/CatfProd-02-03-

2017.pdf.

United States Department of Agriculture/National Agricultural Statistics Service

(USDA/NASS). (2018). Production and Inventory of Channel Catfish. United

States Department of Agriculture, Washington, DC,

http://usda.mannlib.cornell.edu/usda/nass/CatfProd//2010s/2018/CatfProd-02-02-

2018.pdf.

Veeruraj, A., Arumugam, M., Ajithkumar, T., & Balasubramanian, T. (2015). Isolation

and characterization of collagen from the outer skin of squid (Doryteuthis

singhalensis). Food Hydrocolloids, 43, 708-716.

Veeruraj, A., Arumugam, M., & Balasubramanian, T. (2013). Isolation and

characterization of thermostable collagen from the marine eel-fish (Evenchelys

macrura). Process Biochemistry, 48, 1592-1602.

Villamonte, G., Jury, V., Jung, S., & de Lamballerie, M. (2015). Influence of xanthan

gum on the structural characteristics of myofibrillar proteins treated by high

pressure. Journal of Food Science, 80, C522-C531.

Vojdani, F. (1996). Solubility. Methods of testing protein functionality, 11-60.

Wright, D. J., Leach, I. B., & Wilding, P. (1977). Differential scanning calorimetric

studies of muscle and its constituent proteins. Journal of the Science of Food and

Agriculture, 28, 557-564.

Wu, J., & Ding, X. (2002). Characterization of inhibition and stability of soy-protein-

derived angiotensin I-converting enzyme inhibitory peptides. Food research

international, 35, 367-375.

Wu, Q., Du, J., Jia, J., & Kuang, C. (2016). Production of ACE inhibitory peptides from

sweet sorghum grain protein using alcalase: Hydrolysis kinetic, purification and

molecular docking study. Food Chemistry, 199, 140-149.

Xiong, G., Cheng, W., Ye, L., Du, X., Zhou, M., Lin, R., Geng, S., Chen, M., Corke, H.,

& Cai, Y.-Z. (2009). Effects of konjac glucomannan on physicochemical

properties of myofibrillar protein and surimi gels from grass carp

(Ctenopharyngodon idella). Food Chemistry, 116, 413-418.

Xu, J., Mukherjee, D., & Chang, S. K. (2018). Physicochemical properties and storage

stability of soybean protein nanoemulsions prepared by ultra-high pressure

homogenization. Food Chemistry, 240, 1005-1013.

204

Yang, S.-Y., Lee, S., Pyo, M. C., Jeon, H., Kim, Y., & Lee, K.-W. (2017). Improved

physicochemical properties and hepatic protection of Maillard reaction products

derived from fish protein hydrolysates and ribose. Food Chemistry, 221, 1979-

1988.

Yasui, T., & Samejima, K. (1990). Recent advances in meat science in Japan:

Functionality of muscle proteins in gelation mechanism of structured meat

products. Japan Agricultural Research Quarterly, 24, 131-140.

Yata, M., Yoshida, C., Fujisawa, S., Mizuta, S., & Yoshinaka, R. (2001). Identification

and characterization of molecular species of collagen in fish skin. Journal of Food

Science, 66, 247-251.

Yates, L., Dutson, T., Caldwell, J., & Carpenter, Z. (1983). Effect of temperature and pH

on the post-mortem degradation of myofibrillar proteins. Meat Science, 9, 157-

179.

Yin, T., Reed, Z. H., & Park, J. W. (2014). Gelling properties of surimi as affected by the

particle size of fish bone. LWT-Food Science and Technology, 58, 412-416.

Zambrowicz, A., Pokora, M., Setner, B., Dąbrowska, A., Szołtysik, M., Babij, K.,

Szewczuk, Z., Trziszka, T., Lubec, G. and Chrzanowska, J., (2015).

Multifunctional peptides derived from an egg yolk protein hydrolysate: isolation

and characterization. Amino Acids, 47, 369-380.

Zhang, T., Xu, X., Li, Z., Wang, Y., Xue, Y., & Xue, C. (2018). Interactions and phase

behaviors in mixed solutions of κ-carrageenan and myofibrillar protein extracted

from Alaska Pollock surimi. Food Research International, 105, 821-827.

Zhang, T., Xue, Y., Li, Z., Wang, Y., & Xue, C. (2015). Effects of deacetylation of

konjac glucomannan on Alaska Pollock surimi gels subjected to high-temperature

(120° C) treatment. Food Hydrocolloids, 43, 125-131.

Zhang, Y., Liu, W., Li, G., Shi, B., Miao, Y., & Wu, X. (2007). Isolation and partial

characterization of pepsin-soluble collagen from the skin of grass carp

(Ctenopharyngodon idella). Food Chemistry, 103, 906-912.

Zhang, W. W., Lu, J. F., Jiao, D. L., Lin, L., Weng, S. B., &amp; Jiang, S. T. (2009).

Extraction of Fish Oil from Viscera of Channel Catfish through Potassium-

method. Journal of Food Science (Chinese), 24, 007.

Zhang, Y., Pechan, T., & Chang, S. K. (2018). Antioxidant and angiotensin-I converting

enzyme inhibitory activities of phenolic extracts and fractions derived from three

phenolic-rich legume varieties. Journal of Functional Foods, 42, 289-297.

205

Zhou, C., Li, Y., Yu, X., Yang, H., Ma, H., Yagoub, A. E. A., Cheng, Y., Hu, J., & Otu,

P. N. Y. (2016). Extraction and characterization of chicken feet soluble collagen.

LWT-Food Science and Technology, 74, 145-153.