physico-biochemical properties of extruded cape horse ...epubs.surrey.ac.uk › 852198 › 1 ›...

i

Physico-biochemical properties of extruded Cape

Horse mackerel (Trachurus capensis)

By

Stephen Carmelo Barrion

A thesis submitted to the University of Surrey in partial fulfilment of the requirements

for the Degree of Doctor of Philosophy in the Faculty of Health and Medical Science

Department of Nutritional Sciences

Faculty of Health and Medical Sciences

University of Surrey

Guildford, Surrey GU2 7XH

September 2018

ii

Declaration of originality

This thesis and the work to which it refers are the results of my own efforts. Any ideas, data,

images or text resulting from the work of others (whether published or unpublished) are fully

identified as such within the work and attributed to their originator in the text, bibliography

or in footnotes. This thesis has not been submitted in whole or in part for any other academic

degree or professional qualification. I agree that the University has the right to submit my

work to the plagiarism detection service TurnitinUK for originality checks. Whether or not

drafts have been so-assessed, the University reserves the right to require an electronic

version of the final document (as submitted) for assessment as above.

iii

ABSTRACT

Cape horse mackerel (Trachurus capensis) is an underutilized pelagic fish from Namibia. In

this study, the physico-chemical characteristics, nutritional composition and functionality

(thermal and rheological) properties of horse mackerel and local cereals found in Namibia

were investigated. This knowledge of the physico-chemical composition and functionality of

these ingredients and their behaviour in mixed food model systems was applied to optimize

the extrusion process, used to make snack foods. Proximate, biochemical, amino acid and

fatty acid analyses, and differential scanning calorimetry (DSC), rheology and Raman

spectroscopy studies were conducted on horse mackerel mince, cassava, maize, pearl millet

flour, mixtures of these ingredients, an extruded horse mackerel snack and a commercial

shrimp cracker. Differences in the protein bands identified by Raman spectroscopy provided

detailed information on the protein structure of the raw materials, protein denaturation and

gel formation of the extruded snacks. Products of similar textural properties were developed

despite the varying composition of mixtures. A mixture comprising 20% horse mackerel,

20% maize flour, 30% pearl millet flour and 30% cassava flour showed the optimum starch

and protein interactions as evidenced by thermal and rheological properties; this formulation

was used to develop the horse mackerel extruded snack. Extrusion provided an efficient

method for combining essential protein and lipids from fish and starchy cereal mixtures into a

balanced and nutritious snack food. The horse mackerel extruded snack product had a higher

protein quality compared to the commercial shrimp cracker. The raw material quality used in

the development of an extruded horse mackerel snack significantly influenced its final quality

and total cost. Poor quality fish used in the processing of the extruded horse mackerel snack

resulted in an increase in failure costs and consequently the total quality cost. Failure costs

decreased from 80% to 40% of Total Quality Costs when good raw material was used.

iv

ACKNOWLEDGEMENTS

I would like to acknowledge and thank the following individuals and institutions for their

assistance and support which made the study possible.

My special thanks and gratitude go to Prof. Nazlin Howell for her expertise, guidance and

productive comments. I am especially grateful to her reading, reviewing the thesis and

providing feedback promptly. Her generous moral support and unceasing encouragement

motivated me to carry on and finish my study. It was a true privilege to work with her, in her

capacity as the Project Leader of the Securefish Project and as the Main Supervisor of this

Doctoral Thesis.

I would like to thank Mr. Chinemba Samundengu and Dr. Komeine Nantanga, my co-

supervisors, for their valuable contribution. Their unfailing feedback, cooperation and

friendship helped me overcome obstacles I experienced during my research.

My appreciation also goes to the following people and institutions:

Dr. Farah Badii, Dr. Marwa Yousr and Dr. Pooimun Leong for their valuable assistance they

gave me at the University of Surrey laboratories.

Prof. Cheow Chong Seng (deceased) the Securefish Work Package leader whose extrusion

work provided inspiration for this research.

v

The Securefish Project for their financial support and funding of this research. I acknowledge

Dr. Samuel Mafwila the local Project Co-ordinator of the Securefish Project at the University

of Namibia.

Colleagues and friends at the University of Namibia, especially, Dr Erick Kandiwa for their

interest in my work.

The University of Namibia’s Staff Development Fellowship program for providing some

financial assistance for my PhD studies.

Prof. Timothy Rennie, Mr. Anthony Ishola, Ms. Monde Lusepani, Ms. Valerie Aluteni and

Ms. Muhuulu for the use of the analytical equipment and facilities at the School of Pharmacy

of the University of Namibia.

I am very grateful to my wife Gisela Ymata Barrion, daughter Gien Carmela Barrion, mother,

Estefania Barrion, sisters, Carrie, Irene and brothers-in-law, Wilfred and Ernest and the entire

Barrion, Quitasol, Schmidt, Ymata and Garthoff families for their understanding, love,

support and prayers.

I am very thankful to Uncle Tony Quitasol, Auntie Nida Quitasol and cousin Toni-Louise

Quitasol for their love and hospitality during my visits to the United Kingdom.

I would also like to remember my deceased father, Irineo Barrion, whose inspiration made

me persist to carry on. Above all I am most grateful to God our Almighty Father.

vi

ABBREVIATIONS

Ai Appraisal costs

AM Actomyosin

AOAC Association of Official Analytical Chemists

BSA Bovine Serum Albumin

C-H Carbon-hydrogen

CH2 Methylene group

CH3 Methyl group

CSC Commercial shrimp cracker

DHA Docosahexaenoic acid

DLO Dienst Landbouwkundig Onderzoek

DSC Differential Scanning Calorimetry

Dwb Dry weight basis

EPA Eicosapentaenoic acid

FHMEES Fried horse mackerel extruded expanded snack

G’ Storage modulus

G” Loss modulus

HCL Hydrochloric acid

HPLC High Performance Liquid Chromatography

I855/I830 Ratio of tyrosine vibrational residues

vii

KCl Potassium chloride

kDa Kilo Dalton

Mr Relative Molecular mass

MUFA Monounsaturated fatty acids

TMAO Trimethylamine oxide

TQC Total Quality Cost

TQM Total Quality Model

PCB Polychlorinated biphenyls

Pi Prevention costs

PITC Phenyl isothiocyanate

PUFA Polyunsaturated fatty acids

PV Peroxide value

QCM Quality Cost Model

QIM Quality Index Method

SFA Saturated fatty acids

TBA Thiobarbituric acid

TBARS Thiobarbituric acid reactive species

TEA Triethylamine

TVBN Total Volatile Basic Nitrogen

UHME Uncooked horse mackerel extruded snack

viii

LIST OF TABLES

1.1 Proximate and chemical composition (g /100 g) of pearl millet,

maize and cassava 20

1.2 Amino acid content of processed cassava, pearl millet and maize 24

2.1 Proximate and chemical composition (g /100 g) of horse

mackerel (Trachurus capensis) 41

2.2 Concentration of myofibrillar and sarcoplasmic proteins in horse mackerel 43

2.3 Amino acid composition of horse mackerel 45

2.4 Fatty acid profile of horse mackerel mince 46

2.5 Proximate and chemical composition of pearl millet, maize and cassava flour 47

2.6 Amino acid compositions of Namibian cereal flours 49

3.1 Storage modulus (G’Pa) and loss modulus (G”Pa) of Horse mackerel

mince, pearl millet, maize and cassava flours 58

3.2 Transition temperatures and enthalpy change for pearl millet, maize,

cassava and horse mackerel mince 63

3.3 Relative peak intensity of Raman spectra for regions 3200-700cm-1

for Horse mackerel mince, Pearl millet flour, Maize flour and

Cassava flour 64

4.1 Amino acid compositions of Composite Flour (without fish) and

Composite Flour (with fish) 74

ix

4.2 Transition temperatures and enthalpy change for Composite Flour

(without fish) and Composite Flour (with fish) 76

4.3 Storage modulus (G’Pa) and loss modulus (G”Pa) of Composite Flour

(without fish)and Composite Flour (with fish) 80

4.4 Relative peak intensity of Raman spectra for regions 3200-700 cm-1 for

Composite Flour (without fish) and Composite Flour (with fish) 81

4.5 Composition of composite flours 83

4.6 Transition temperatures and enthalpy change for composite flours with varying

proportions of horse mackerel mince (10%,20%,30%) and pearl millet flour

(40%, 30%, 20%) 85

4.7 Storage modulus (G’Pa) and loss modulus (G”Pa) of composite flours with

varying proportions of horse mackerel mince (10%,20%,30%) and pearl

millet flour (40%, 30%, 20%) 89

5.1 Die swell and expansion (%) of the three product composite flours (before

and after extrusion) compared to a commercial shrimp cracker 100

5.2 Proximate and chemical composition (g /100 g) of uncooked Horse

mackerel extruded snack and fried commercial shrimp cracker (CSC) 101

5.3 Amino acid composition of uncooked horse mackerel extruded

snack (UHME) and commercial shrimp cracker (CSC) 102

x

5.4 Fatty acid profile of the fried horse mackerel extruded expanded snack

(FHMEES) 107

5.5 Fatty acid composition of sunflower oil (Helianthus annuus) 108

5.6 Texture analyses of uncooked Horse mackerel non-expanded extruded

snack made from Composite flour 2, uncooked commercial shrimp cracker,

fried horse mackerel extruded expanded snack made from Composite flour 2

and fried commercial shrimp cracker 112

5.7 Transition temperatures and enthalpy change for the uncooked

horse mackerel extruded snack (UHMES) made from Composite flour 2

and shrimp crackers 114

5.8 Relative peak intensity of Raman spectra for regions 3200-700 cm-1

for uncooked horse mackerel non-expanded extruded snack and

commercial prawn cracker 115

6.1 Quality Index Method scheme (adapted from Larson et al., 1992 ) 123

6.2 Quality, market and production parameters for extruded products 125

6.3 Relationships used to estimate quality costs components for extruded

horse mackerel 126

6.4 Quality, market and production parameters for extruded products 130

xi

LIST OF FIGURES

1.1 Cape horse mackerel (Trachurus capensis) 3

1.2 Structure of myosin and actin and the actomyosin complex 10

1.3 Muscle contraction and relaxation 11

1.4 Collagen conversion to gelatin 12

1.5 Actomyosin muscle structure and actomysosin 13

1.6 The basic structure of collagen 16

1.7 Starch granule structure of pearl millet 18

1.8 Single screw extruder operation 26

1.9 Diagram of extrudate expansion 27

2.1 Phast gel electrophoresis separation pattern of sarcoplasmic and

myofibrillar protein in horse mackerel 44

3.1 A dynamic temperature sweep (20-90-20°C) of horse mackerel mince

(100% w/w) 56

3.2 A dynamic temperature sweep (20-90-20°C) of pearl millet flour

(50% w/v in distilled water) 57

3.3 A dynamic temperature sweep (20-90-20°C) of maize flour


xii

3.4 A dynamic temperature sweep (20-90-20°C) of cassava flour


3.5 Typical DSC profiles of cereals pearl millet, maize and cassava


3.6 Typical DSC profile of horse mackerel (100% w/v) 62

4.1 A dynamic temperature sweep (20-90-20°C) of Composite Flour (without fish)

(25:37.5:37.5 maize flour:cassava flour:pearl millet flour) with no

horse mackerel mince 77

4.2 A dynamic temperature sweep (20-90-20°C) of Composite Flour (with fish)

(20:20:30:30 horse mackerel:maize:cassava: pearl millet) with

horse mackerel mince 79

4.3 A dynamic temperature sweep (20-90-20°C) of Composite flour 1

(10:20:30:40 horse mackerel mince:maize:cassava:pearl millet) 86





5.1 Single screw extruder used to process a horse mackerel extruded snack 95

5.2 Flow diagram of the development of a horse mackerel extruded snack 96

xiii

5.3 Uncooked Commercial Shrimp Cracker and uncooked Horse mackerel

non-expanded extruded snacks made from 3 composite flours 104

5.4 Fried Commercial Shrimp Cracker and Fried Horse mackerel non-expanded

extruded snacks made from 3 composite flours 105

5.5 A dynamic temperature sweep (20-90-20°C) of an uncooked Horse

mackerel non-expanded extruded snack made from Composite flour 2


5.6 A dynamic temperature sweep (20-90-20°C ) of a commercial

shrimp cracker 111

5.7 Texture profile of an uncooked Horse mackerel non-expanded extruded

snack made from Composite flour 2 and Uncooked commercial shrimp

cracker using the TAX-T2 Texture analyser 113

5.8 Fried Horse mackerel non-expanded extruded snack made from Composite

flour 2 and a fried commercial shrimp cracker using the TAX-T2

Texture analyser 113

6.1 Influence of raw material quality on product quality in extruded Cape

horse mackerel 128

6.2 Variation of filleting yield with raw material quality, dimensionless, for

horse mackerel 129

xiv

6.3. Total quality costs, failure costs and controllable costs as a function of

product quality for the extruded horse mackerel fish snack 132

xv

TABLE OF CONTENTS

Declaration of originality ........................................................................................................ ii

ABSTRACT ............................................................................................................................ iii

ACKNOWLEDGEMENTS ................................................................................................... iv

ABBREVIATIONS ................................................................................................................. vi

LIST OF TABLES ............................................................................................................... viii

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

CHAPTER 1 ............................................................................................................................. 1

1. GENERAL INTRODUCTION ........................................................................................... 1

1.1 Food supplies .................................................................................................................. 1

1.2 The Namibian fishing industry ..................................................................................... 1

1.2.1 Biological characteristics of Cape Horse Mackerel ................................................. 2

1.3 Protein structure and properties .................................................................................. 3

1.3.1 Primary, secondary, tertiary and quaternary structure ......................................... 3

1.3.2 Physical–chemical properties of proteins .................................................................. 5

1.3.3 Covalent and non-covalent bonding .......................................................................... 6

1.3.4 Functional properties of proteins .............................................................................. 6

1.3.4.1 Gelation ..................................................................................................................... 6

1.3.4.2 Foaming and emulsification .................................................................................... 8

1.3.4.3 Raman Spectroscopy ................................................................................................ 8

1.4 Composition and properties of fish protein ................................................................. 9

xvi

1.4.1 Myofibrillar proteins ............................................................................................... 9

1.4.2 Sarcoplasmic Proteins ........................................................................................... 14

1.4.3 Myoglobin and Haemoglobin ............................................................................... 14

1.4.4 Nonprotein Nitrogenous Constituents ................................................................. 15

1.4.5 Connective Tissue Proteins ................................................................................... 15

1.5 Lipids ......................................................................................................................... 17

1.6 Cereals ....................................................................................................................... 18

1.6.1 Nutritional composition of pearl millet ............................................................... 18

1.6.2 Nutritional composition of maize ......................................................................... 21

1.6.3 Nutritional composition of Cassava ..................................................................... 23

1.7 Extrusion ................................................................................................................... 25

1.7.1 Factors affecting the extrusion process ............................................................... 27

1.8 OBJECTIVES OF STUDY ............................................................................................. 28

CHAPTER 2 ........................................................................................................................... 30

2. PROXIMATE AND BIOCHEMICAL COMPOSITION OF THE CAPE HORSE

MACKEREL (TRACHURUS CAPENSIS), PEARL MILLET FLOUR, MAIZE AND

CASSAVA ............................................................................................................................... 31

2.1 Introduction .................................................................................................................. 31

2.2 Materials and Methods ................................................................................................ 31

2.2.1 Materials ................................................................................................................. 31

2.2.2 Methods ...................................................................................................................... 32

2.2.2.1 Moisture content ................................................................................................. 33

xvii

2.2.2.2 Ash content .......................................................................................................... 33

2.2.2.3 Crude fat content ................................................................................................ 34

2.2.2.4 Fatty acid analysis .............................................................................................. 35

2.2.2.5 Protein determination ........................................................................................ 36

2.2.2.6 Determination of amino acids ............................................................................ 37

2.2.2.7 Derivatisation of amino acids with phenylisothiocyanate PITC .................... 37

2.2.2.8 Protein extraction .............................................................................................. 38

2.2.2.9 Bradford Protein assay ...................................................................................... 39

2.2.2.10 Phast gel electrophoresis .................................................................................. 39

2.4 Statistical analyses ........................................................................................................ 40

2.5 Results and discussion .............................................................................................. 41

2.6 Conclusion ..................................................................................................................... 50

CHAPTER 3 ........................................................................................................................... 51

3. RHEOLOGICAL, THERMODYNAMIC (DSC) AND STRUCTURAL (FT- RAMAN

SPECTROSCOPY) STUDIES OF FISH MINCE, PEARL MILLET, MAIZE FLOUR

AND CASSAVA STARCH RAW MATERIALS .............................................................. 52

3.1 Introduction .................................................................................................................. 52


3.2.1 Small-deformation oscillatory measurements ..................................................... 53

3.2.2 Temperature sweeps .............................................................................................. 54

3.2.3 Differential scanning calorimetry (DSC) Measurements................................... 54

3.2.4 FT-Raman spectroscopy ....................................................................................... 55

xviii

3.3 Results and discussion .................................................................................................. 55

3.3.1 Rheological studies ................................................................................................ 55

3.3.2 DSC Studies ............................................................................................................ 61


3.4 Conclusion ..................................................................................................................... 68

CHAPTER 4 ........................................................................................................................... 69

4. PRODUCT DEVELOPMENT AND PRODUCT MIXTURES .................................... 70

4.1 Introduction .................................................................................................................. 70


4.2.1 Determination of amino acids ............................................................................... 72

4.2.2 Small-deformation oscillatory measurements ..................................................... 72

4.2.3 Temperature sweeps .............................................................................................. 72

4.2.4 Differential scanning calorimetry (DSC) measurements ................................... 72




4.4.1.1 Amino acid analysis of mixture of flours and fish ........................................... 73

4.4.1.2 Differential scanning calorimetry ..................................................................... 75

4.4.1.3 Small deformation rheology .............................................................................. 76

4.4.1.4 Raman Spectroscopy .......................................................................................... 80

4.4.2 Cereal (maize and cassava) composite flours with varying proportions of horse

mackerel mince and pearl millet flour (Composite flour 1, 2, and 3)............................ 83

xix

4.4.2.1 Differential Scanning Colorimetry of Composite flour 1,2 and 3 .................. 83

4.4.2.2 Small deformation rheology .............................................................................. 86

4.5 Conclusion ..................................................................................................................... 90

CHAPTER 5 ........................................................................................................................... 91

5. NUTRITIONAL AND BIOCHEMICAL PROPERTIES OF EXTRUDED CAPE

HORSE MACKEREL (TRACHURUS CAPENSIS), PEARL MILLET FLOUR, MAIZE

AND CASSAVA PRODUCTS .............................................................................................. 92

5.1 Introduction .................................................................................................................. 92


5.2.1 Materials .................................................................................................................... 94

5.2.2 Methods ...................................................................................................................... 94

5.2.2.1 Laboratory-scale single-screw extruder ........................................................... 94

5.2.2.2 Moisture content ................................................................................................. 97

5.2.2.3 Protein determination ............................................................................................ 97

5.2.2.4 Fat content ........................................................................................................... 97

5.2.2.5 Ash content .......................................................................................................... 97

5.3 Biochemical characterization ...................................................................................... 98

5.3.1 Fatty acid analysis ................................................................................................. 98

5.3.2 Amino acid analyses .............................................................................................. 98

5.3.2.1 Determination of amino acids ............................................................................ 98

5.3.2.2 Derivatisation of amino acids with phenylisothiocyanate (PITC).................. 98

5.3.3 Die swell .................................................................................................................. 98

xx

5.3.4 Expansion Ratio ........................................................................................................ 98

5.3.5 Texture analyses .................................................................................................... 99



5.6 Conclusion ................................................................................................................... 118

CHAPTER 6 ......................................................................................................................... 120

6. INFLUENCE OF RAW MATERIAL ON THE QUALITY OF THE EXTRUDED

HORSE MACKEREL SNACK AND A QUALITY COST MODEL ............................. 121

6.1 Introduction .................................................................................................................... 121

6.2 Materials and Methods .............................................................................................. 122

6.2.1 Materials ............................................................................................................... 122

6.2.2 Methods .................................................................................................................... 122

6.2.2.1 Fish sample preparation .................................................................................. 122

6.2.2.2 Yield ................................................................................................................... 123

6.2.2.3 Sensory Evaluation of raw material with Quality Index Method ................ 123

6.2.2.4 Development of the horse mackerel fish-based snack ................................... 124

6.2.2.5 Sensory evaluation of the final fish-based snack developed product ........... 124

6.2.2.6 Model and assumptions .................................................................................... 125

6.2.2.7 Statistical analyses ............................................................................................ 127

6.3 Results and discussion ................................................................................................ 128

6.3.1 Influence of raw material quality ....................................................................... 128

6.3.1.1 Extruded product quality ................................................................................ 128

xxi

6.3.2 Filleting yield and productivity .......................................................................... 129

6.3.3 Quality, market and production parameters .................................................... 130

6.3.4 Quality costs estimation ...................................................................................... 131

6.3.5 Quality cost for extruded fish product .............................................................. 131

6.4 Conclusions ................................................................................................................. 133

CHAPTER 7 ......................................................................................................................... 134

7. GENERAL DISCUSSION .............................................................................................. 135

7.1 Proximate and biochemical composition of Cape horse mackerel (Trachurus

capensis), pearl millet (Pennisetum glaucum L. Br, Maize (Zea mays) and Cassava

(Manihot esculenta) .............................................................................................................. 136

7.2 Rheological, DSC, FT-Raman spectroscopy of horse mackerel fish mince, pearl

millet, maize, cassava flours, product development and mixtures .............................. 138

7.3 Product Development and Mixtures ......................................................................... 139

7.4 Extrusion ..................................................................................................................... 141

7.5 Influence of raw material on the quality of fish products ...................................... 143

7.6 CONCLUSIONS ............................................................................................................ 144

7.7 Further work .............................................................................................................. 147

8. REFERENCES ................................................................................................................. 150

9. APPENDICES .................................................................................................................. 171

1

CHAPTER 1

1. GENERAL INTRODUCTION

1.1 Food supplies

In the next four decades, the demand for food will increase due to an increase in population

and the need to meet their consumptions needs (Godfray et al, 2010). In order to address

these needs, global food production should increase by 70 to 100% to meet the demand for

food by 2050 (Tomlinson, 2013). Land, water, energy and fishery resources are increasingly

being competed for (Godfray et al, 2010). The challenges on the food system can be

addressed by reducing waste and producing more food efficiently (Godfray and Garnett,

2014). The sub-Saharan region currently has a population of 900 million and it is envisaged

that the population in Sub-Sahara will double by 2050 (Josephson, et al., 2014). Food security

is greatest at risk in the Sub-Saharan African region which relies substantially in the

importation of food (van Ittersum et al., 2016). In Namibia, fifty percent of the food that is

consumed domestically is imported (Bruntrup and Herrmann, 2012). Therefore, there is a

need to produce more food within the country or utilise local food resources currently

available including marine and aqua cultured fish.

1.2 The Namibian fishing industry

The Benguela ecosystem includes the entire Namibian coast, west coast of South Africa and

southern Angola (Boyer and Hampton, 2001). The coastal upwelling in the Benguela current

results in the production of plankton which makes these waters rich in pelagic e.g. horse

2

mackerel and demersal fish. The Cape horse mackerel (Trachurus capensis) is a popular fish

species caught in the northern part of the Benguela waters, which is characterized by cool

surface waters and high biological productivity (Sakko, 1998). In Namibia, midwater

trawling is the preferred method for harvesting pelagic fish species, particularly adult size

horse mackerel. Purse seiners, on the other hand, are used to catch juveniles which are mostly

processed into fish feed and oil (Boyer and Hampton, 2001). The average total annual catch

of Horse mackerel over the past four decades was approximately 300 000 tons (Bauleth-

D’Almeida, Krakstad and Kanandjembo, 2001).

According to the Food and Agriculture Organization (FAO), horse mackerel has

approximately 50% edible flesh (FAO, 1989) i.e. a yield of 50%. The edible portion has a

protein composition of 19.7% and a fat content of 3.9%. The inedible portion comprising the

bones have been reported to have a protein content of 538 g/kg and 509 g/kg lipid content

(Toppe et al., 2007).

In Namibia, the unemployment rate is at 34% and high rates of poverty have led to an

estimated 42.3% of the population to be undernourished (WFP, 2017). Thus, horse mackerel

as a good source of protein could partly address this nutritional problem.

1.2.1 Biological characteristics of Cape Horse Mackerel

Cape horse mackerel has an elongate body (Akawa and Nashima, 2014) covered with small

scales. It has a large pointed head, a big mouth with numerous small teeth (Figure 1.1) (FAO,

2013).

3

Figure 1.1. Cape horse mackerel (Trachurus capensis) (Castelnau, 1861) (FAO, 2013)

The lower part of the upper jaw extends to the bottom regions of the eye, while the bottom

jaw protrudes slightly with a well-developed eyelid. Two small nostrils are situated closely

together; the top nostril is ovate in shape and bottom nostril moon shaped. Cape horse

mackerel can grow up to 60 cm in length, but the common length of horse mackerel is 30 cm.

The colour of the top side of the body and upper section of the head is grey-black with an off

blue green colour. The remaining bottom side of the body and head are a pale, white to shiny

colour. Like most Trachurus species, horse mackerel typically feeds on zooplankton and

small fish (Yankova, Raykov and Frateva, 2008).

1.3 Protein structure and properties

1.3.1 Primary, secondary, tertiary and quaternary structure

Proteins consist of polypeptide polymer chains comprising repeating units of amino acids

varying in characteristics (Yüksektepe et al, 2008). Amino acids consist of a carbon atom

(Cα) and an amino group (NH2) (Prošková, 2014). Attached to the carbon atom of the amino

4

acid, is an arrangement of a hydrogen atom (H), a carboxyl group (COOH) and a side chain

(R) that gives the amino acid its characteristic physical-chemical properties. The amino acid

proline has a side chain that is bonded to a nitrogen making it a cyclic imino acid. (Creighton,

1993) The protein characteristics also depend on the sequence of amino acids in the

polypeptide chain. The carboxyl group of one amino acid is linked to the amino group of

another amino acid to form a peptide bond, with the removal of water. There are about 20

amino acids that are commonly present in proteins, making them very complex polymers.

The simplest amino acid is glycine with an H for the side chain. The aliphatic amino acids

comprise Ala, Val, Leu and Ile. Serine and threonine are aliphatic but also polar hydroxyl

groups. Acidic residues consist of aspartic and glutamic acids which also exist as amide

forms asparagine and glutamine. The basic residues are positively charged lysine and

arginine. The imidazole group of histidine is very reactive. The aromatic residues

phenylalanine, tyrosine and tryptophan show ultraviolet and fluorescence absorbance

properties and are important for protein interactions. Methionine and cysteine contain sulphur

groups and are involved in disulphide-sulphydryl interchange processes that affect gelation.

A typical protein molecule consists of 200 to 300 amino acids; however, these can increase

up to 30 000 amino acids on a single chain. In certain proteins, the arrangement of a single

polypeptide chain in its three-dimensional structure determines the final protein created

(Emerson and Amala, 2017).

Proteins comprise four structural levels including primary, secondary, tertiary and quaternary

structures (Han et al., 2018). The linear sequence of amino acids linked by covalent peptide

bonds is known as the primary structure. The linear nature has the possibility of adopting a

three-dimensional regular conformation. The secondary structure refers to the arrangement of

5

a polypeptide into α-helices, β-strands and random coil. The righthanded α-helix is the most

common structure. The carbonyl oxygen of each amino acid hydrogen bonds to the backbone

NH of every fourth residue. The side chains protrude outwards (Creighton, 1993). The

second secondary structure comprises the β-strands which are stable as the β-sheet, a ribbon–

like structure that allows intra- and inter-molecular interactions. The tertiary structure relates

to a folding of the polypeptide chain into a small globular structure like that of serum albumin

and ovalbumin, that is generally water-soluble. Tertiary structures can also be fibrous, for

example the coiled coil structures and triple helix structure of collagen and gelatin. The triple

helices can be arranged next to each other to form microfibrils of collagen. Finally, the

quaternary structure refers to the association of two or more polypeptide chains, for example,

the four globin polypeptide chains surrounding the heme in haemoglobin or myoglobin

(Howell, 1992).

1.3.2 Physical–chemical properties of proteins

Low heating temperatures (36-40°C) for a short time can cause substantial conformational

changes in proteins (Tornberg, 2005). In contrast, high temperatures and longer times are

reported to result in lowered interaction between different proteins. The exposure of the

amino acid reactive groups results from conformational changes leading to protein

denaturation. Protein denaturation is due to the exposure of hydrophobic, sulphydryl and

other reactive groups which can participate in protein interactions (Howell, 2000). A

reversible gel with non-covalent bonds is formed. Thereafter, at temperatures above 80°C. a

covalently bonded non-reversible gel is formed with most food proteins, except gelatin which

is thermoreversible. Intramolecular forces influence these protein structural conformations,

particularly the forces affecting the strength and length of the chemical bond between two

atoms such as a C-C bond (Howell, 1992). The pH sensitivity of protein molecules relate to

its isoelectric point (pI) and the pH of the medium.

6

1.3.3 Covalent and non-covalent bonding

In the primary structure of proteins covalent bonds including the peptide bonds exist. During

protein denaturation the peptide bonds remain intact (Howell, 1992).Thus, only secondary

and tertiary structures are affected. The peptide bonds in protein primary structures are only

broken down through hydrolysis in strong acid or alkali environments or by proteolytic

enzymes (Kristinsson and Rasco, 2000). In the tertiary structure covalent disulphide bonds

which are heat sensitive are present. Therefore, protein denaturation occurs through thermal

application for example during heat gelation or with shear as in foaming.

1.3.4 Functional properties of proteins

Proteins influence the functional properties of food quality and stability (Han et al., 2018).

The nutritional and sensorial quality of manufactured and stored food are due to various

protein-protein interactions (Howell, 2000). The functional properties of proteins are

influenced by the molecular forces involved in protein interactions (Howell, 1992; Howell,

2000). During the processing and storage of food these protein interactions cause biochemical

changes in food (Onwulata and Qi, 2006).

1.3.4.1 Gelation

During heating, changes in the behaviour of proteins relates to aggregation and

intermolecular interaction (Howell, 1992). Fish protein aggregation can result from long term

frozen storage (Howell, 2000). Changes in the protein structure and protein interactions of

smooth and skeletal muscles proteins alter the protein side chains enhancing gelation

properties. Skeletal muscles have better gelling properties than smooth muscle myosin due to

differences in ionizable and hydrophobic residues (Howell, 2000). In smooth muscle myosin,

7

glutamic acid disrupts the water structure near the hydrophobic groups resulting in weak

gelation (Badii and Howell, 2006).

Rheology provides quantitative measurements regarding protein behaviour. Protein size and

shape affects viscosity. Protein-protein interactions are observed through the measurement of

resistance to flow or viscosity prior to gel formation. Heating of globular proteins at high

temperatures causes conformational changes which can be measured through changes in

viscosity and through differential scanning calorimetry (Tornberg, 2005).

Protein gels comprise three-dimensional networks with water bound tightly to polar groups of

polypeptide chains or less to network interstitial spaces (Howell, 1992). Protein gelation

reactions depend on protein concentration and the careful application of heating temperature

over a specific time. At low temperatures, globular proteins are partially unfolded and at high

temperatures or at the isoelectric point or pI, aggregates are formed. Gel strength relates to

the cohesiveness of protein interactions. Moreover, gel strength depends on synergistic

interactions between proteins and other proteins or polysaccharides that influence viscosity

and gelation (Howell, 2000). Hydrophobic interactions and the functionality of proteins affect

the heat gelation of proteins and gel strength of heated gels on cooling. Protein interactions

depend on the species, type of protein, heating time and temperature and affect the partial

unfolding of the protein molecules to end in a ‘molten state’ (Howell, 2000). Differences in

the N-terminal and C-terminal of amino acids and conformational changes in the primary

structure may also affect gelation (Howell,1992). Gelation is dependent on protein

compatibility with other components during heating (Baeza et al. 2002).

8

1.3.4.2 Foaming and emulsification

The application of chromatographic characterization of protein, differential scanning

calorimetry and conformational changes is applicable to the foaming and emulsification

properties of food as it involves dilute protein solutions. These methods are not applicable to

gelation involving large concentration of proteins and rheometers or texture analysers are

used for characterisation. Chemical modifications of proteins that block the exposed reactive

groups allows for the determination of the chemical bonds involved e.g. the reduction of

disulphide bonds by cysteine hydrochloride (Howell, 1992). The shortcoming is the inability

to relate disulphide groups of individual proteins that affect gelation, particularly in mixtures.

1.3.4.3 Raman Spectroscopy

Raman spectroscopy has been applied to study protein and hydrophobic interactions (Howell,

2000). In gels, pastes and solids, Raman spectroscopy was used to determine the presence of

electrostatic interactions. At 450-1900 cm-1 regions both electrostatic and hydrophobic

interactions were observed. These relate to changes in hydrophobic groups, amino acids,

peptides and proteins. Spectral bands CH (1336 cm-1), CH2 (1456 cm-1) and C-H stretching

bands in the 2800-3100cm-1 relate to bending vibrations in hydrophobic interactions. A

decrease in intensity of spectral bands of Trp residues (759 and 1336 cm-1) occur in a non-

polar environment (Howell, 2000). During heating at 90°C, individual proteins can form gels

as indicated by changes in the aromatic and aliphatic CH groups, typical of changes in

hydrophobic interactions, as well as disulphide bonds. The main protein band near 2940cm-1

relates to aromatic, aliphatic and charged amino acids. Bands near 3065 cm-1 are assigned to

aromatic residues whereas aliphatic amino acids are assigned to bands near 2880 and 2900

cm-1.

9

Disadvantages of Raman spectroscopy relates to the interpretation of the C-H stretching

region as the spectrum is very broad. Vibrations of the methylene (CH2) and methyl (CH3)

groups of saturated alkyl groups should correlate with vibrational frequencies and -CH

groups of unsaturated aromatic groups. The hydrophobic interactions of aliphatic and

aromatic residues of proteins relates to the C-H stretching region. The various CH, CH2 and

CH3 groups on the side chains, ionization state and the microenvironment highlights the

complexity of the Raman spectra of amino acids.

1.4 Composition and properties of fish protein

1.4.1 Myofibrillar proteins

Fish are important sources of protein (Steiner-Asiedu, Lied, Lie, Nilsen and Julshamn, 1993).

The structural proteins in fish such as actin and myosin (Figure 1.2) are responsible for

muscle movement (Yildiz, 2009), whereas the myofibrillar proteins with a regulatory

function include tropomyosin, troponin, and actinin (Venugopal and Shahidi, 1996).

10

Figure 1.2. Structure of myosin and actin and the actomyosin complex (Guo and

Greaser, 2017)

Myosin is the major constituent of myofibrils and is soluble in 0.45 M KCl or NaCl solution

(Martone, Busconi, Folco, Traco and Sanchez, 1986). These proteins affect the ability of fish

muscle to form gels at low temperature temperatures (Chan, Gill and Paulson, 1992). The

myosin molecule in horse mackerel consists of two heavy chains (200 kDa) and two light

chains (16 to 20 kDa) (Silva, Mendes, Nunes and Empis, 2006). The myosin head is involved

with ATPase activity (Figure 1.3), which is regulated by calcium and magnesium ions

11

(Venugopal, 2008). G-actin comprises fine elastic ‘S-filaments’ with the function of holding

the actin filaments intact (Eskin, Aliani and Shahidi, 2013).

Figure 1.3. Muscle contraction and relaxation (Guo and Greaser, 2017)

Myosin and actin exist as an actomyosin complex in post-mortem muscles (Li, Xu and Zhou,

2012). Actomyosin (AM) is the primary component of salt-soluble protein in muscles of

animal origin and performs a key function in the texture and manufacturing properties of

meat and the products derived from it (Liu, Zhao, Xiong, Qi and Xie, 2007) which ultimately

contributes to its rheological nature.

12

The primary function of myosin and actomyosin, in fish products is gelation (Gill, Chan,

Phonchareon and Paulson, 1992). Myosin is the main portion (60 to 80%) and has a major

role in the characteristics of gel (Figure 1.4) particularly its heavy chain portion (Ramírez,

Martín-Polo, and Bandman, 2000).

Figure 1.4. Collagen conversion to gelatin (Rolf Kohl, Ludwig Maximilian University of

Munich)

The solubilisation of myofibrils to F-actin and myosin (Figure 1.5), and subsequent

polymerisation to actomyosin results in the formation of gels and increased water holding

capacity (Sun and Holley, 2010). A hydrogel system is created when the myofibrillar protein

gel surrounds water molecules and forms a continuous protein–water cross-linked matrix

(Hwang and Damodaran, 1997).

13

Figure 1.5. Actomyosin muscle structure and actomyosin (Xiong, 2018)

In invertebrates, paramyosin is a major protein found in striated and smooth muscles (Sen,

2005) and is involved in catch contraction of bivalves. Under controlled conditions proteins

can be subjected to an intentional denaturation process in order to exploit their characteristics

and technological functionalities (Yildiz, 2009). In surimi production, the gel-forming ability

of the myofibrillar proteins is exploited. A very strong gel is formed when fish mince is

washed with sodium chloride, which solubilises the salt-soluble myofibrillar proteins

followed by controlled heating and cooling treatments (Tan et al., 2019).

14

1.4.2 Sarcoplasmic Proteins

In fish, myogens and enzymes are part of a large group of proteins, which are soluble in

water or low ionic strength solution that are known as sarcoplasmic proteins (Hemung and

Chin, 2013). Sarcoplasmic proteins constitute between 25 to 30% of fish muscle (Viera et al.,

2018). The majority are found in the cell cytosol and are present in the sarcoplasm (Gaonkar

and McPherson, 2005). Low molecular weight compounds known as albumins are mainly

found in the sarcoplasmic proteins (Shahidi, 1994). These comprise approximately 30% of

the total proteins found in fish muscle (Fauconneau, Grayt and Houlihant, 1995). Generally,

demersal fish have a lower sarcoplasmic protein content than pelagic fish (Sikorski, 2012).

The breakdown of nitrogenous compounds in fish is mainly due to several catalytic enzymes

present in the sarcoplasmic fraction (Shahidi and Botta, 2012).

1.4.3 Myoglobin and Haemoglobin

In fish, myoglobin and haemoglobin are present in high amounts and give muscles its

characteristic red colour due to iron (Pérez-Alvarez and Fernández-López, 2007). These

proteins have a key role in forming complexes such as a heme complex, with approximately

40% available iron (Schönfeldt and Hall, 2011). The proteins comprise one heme per

polypeptide chain (globin), which includes 140 to 160 amino acid residues (Simpson, 2012).

Myoglobin contains one type of polypeptide chain (de Souza and Bonilla-Rodriguez, 2007),

whereas haemoglobins comprise four chains of two different types of polypeptides.

Myoglobin saturated with oxygen is red in colour, oxymyoglobin. During deoxygenation the

colour is reduced to purple-colour (Simpson, 2012).

15

1.4.4 Nonprotein Nitrogenous Constituents

Fish muscle normally has a non-protein nitrogen content higher than that of land animals.

Contents ranging from 10% to 40% have been reported and it includes amino acids

particularly glycine, taurine, alanine and lysine, as well as small peptides, trimethylamine

oxide (TMAO), trimethylamine and nucleotides (Fraser and Sumar, 1998).

1.4.5 Connective Tissue Proteins

In the connective tissue of fish muscles, collagen is a protein found in high amounts,

consisting mostly of type I collagen (Moreno, Jacq, Montero,Gómez-Guillén, Borderías and

Mørkøre, 2016). Collagen comprises approximately 30% of protein (Muyonga, Cole and

Duodu, 2004). Variations in the amino acid composition, especially the imino acids content

(proline, hydroxyproline) is common. Industrial collagen is mostly derived from pig and

bovine skin and bones (Nagaia, Arakib and Suzukia, 2002). The unique properties of fish

collagen allows it to be considered as an alternative to bovine and porcine collagens in the

future (Liu, Li and Guo, 2007). During cooking, collagen is denatured and converted into

gelatin (Nishimoto, Sakamoto, Mizuta and Yoshinaka, 2005). The basic structure of a

collagen molecule (Figure 1.6) are right handed rod-shaped helices comprised of three

parallel α-chains entangled together (Haug, Draget and Smidsrød, 2004).

16

Figure 1.6. The basic structure of collagen (adapted from Badea et al., 2011)

In gelatin, hydrogen bonds stabilize links which are formed between elastic peptide chains.

Fish collagens contain higher levels of essential amino acids than mammalian collagen, with

the exception of hydroxyproline and to a lesser extent proline (Karim and Bhat, 2009). Fish

collagen is less cross-linked and more soluble than the bovine or porcine collagen (Gómez-

Guillén et al., 2011). Cold water fish gelatin has lower melting and gelling points than

porcine and bovine gelatins and therefore result in a weaker gel (Ninan et al.,2014)

17

1.5 Lipids

Lipids occur mostly in membranous areas of fatty fish, while in lean fish, lipids accumulate

in the liver, muscles tissue, and developed gonads (Venugopal, 2008). The total lipid content

is a critical aspect in biological and nutrition studies (Ramalhosa, Paíga, Morais, Alves,

Delerue-Matos and Oliveira, 2012). Consumer awareness of the numerous health benefits of

the inclusion of fish in the human diet in relation to the intake of its high protein and

unsaturated fatty acid contents, has increased (Kaur et al., 2014). Differences in crude lipid

contents of fish muscle may be due to seasonal variation and location (Osako, Kuwahara,

Saito, Hossain and Nozaki, 2003).

Fatty fish species have a high nutritional benefit because they contain substantial amounts of

long chain omega-3 polyunsaturated fatty acids (PUFAs) such as eicosapentaenoic acid

(EPA, C20:5 n-3) and docosahexaenoic acid (DHA, C22:6 n-3) (Farvin, Grejsen and

Jacobsen, 2012). It is believed that the intake of these polyunsaturated fatty acids is beneficial

for health (Eymard, Baron and Jacobsen, 2009). For example, omega-3 PUFA has been

reported to have protective effects against heart disease (Chang and Decklebaum, 2013) and

some types of cancers (Park et al.,2013) and assists in the brain development of children’s

brains (Lauritzen et al., 2016). Many body processes such as inflammation and blood clotting

are regulated by prostaglandins which are made from PUFA. The integrity of membranes of

all living cells are maintained by PUFA. These fats are also required in diets in order to

absorb fat-soluble vitamins A, D, E and K from food (Jabeen and Chaudhry, 2011). In horse

mackerel, it was reported that the n-3 PUFA represented 88–89% of total PUFA (Orban, Di

Lena, Nevigato, Masci, Casinii and Caproni, 2011). The amounts of lipids in horse mackerel

may vary from 2% to 4.5%, depending upon species and individual variation (Aubourg,

Rodriguez and Gallardo, 2005).

18

1.6 Cereals

Cereal grains are generally deficient in lysine, methionine and cysteine (Shewry and Halford,

2002), whereas fish protein is a rich source of protein containing a balanced level of amino

acids (Nehete, Bhambar, Narkhede and Gawali, 2013). Combining fish with cereals is

typically aimed improving the nutritional value of products made from cereal-protein

mixtures.

1.6.1 Nutritional composition of pearl millet

1.6.1.1 Carbohydrates

Starch is the major component of pearl millet and is present in starch granules (Figure 1.7)

within the endosperm of the grain (Annor, Marcone, Bertoft and Seetharaman, 2014). The

largest fraction is rich in starch which is mainly found in the endosperm. It accounts for 75%

to more than 80% of the kernel weight (Hama, Icard-Vernière, Guyot, Picq, Diawara and

Mouquet-Rivier, 2011). The gelatinization temperature of pearl millet starch has been

reported to range between 62.8 to 70.6°C (Annor et al., 2014).

Figure.1.7 Starch granule structure of pearl millet (adapted from Annor et al., 2017)

19

The physico-chemical properties of pearl millet starch are similar to maize. Pearl millet has

gelatinisation temperature that ranges between 60.9 and 78.0°C (Hoover et al., 1996). Maize

has gelatinisation temperatures are between 60 and 62.5°C (Vazquez and Gómez-Aldapa,

2001). In maize, the gelatinisation temperatures can be lower depending on the level of starch

damage or higher (65 to 70°C) if maize starch is modified by nixtamalization. Wheat has

higher gelatinisation temperatures compared to pearl millet that range between 66 and 80°C

(Zeng et al., 2014).

Pasting temperature of starch suspensions ranged from 88.1 to 90.2°C (Bhupender, Rajneesh,

and Baljeet, 2013). The enthalpy of gelatinization of starch granules ranged from 11.8 to 13.2

J/g, and the enthalpy of melting of the retrograded starches was 2.2 to 5.9 J/g. Pearl millet

grain has amylose content of 15.7 to 19.5% (Bhupender et al., 2013). Levels as high as 28.6

to 33.9% starch amylose content have been reported (Annor et al., 2014). Amylose content

can vary greatly in pearl millet starch.

1.6.1.2 Proteins

Most pearl millet cultivars have 13.6 to 15.2% crude protein content (Ali, El Tinay and

Abdalla, 2003; Osman, 2011) as shown in Table 1.1. In pearl millet, approximately 75% of

the total protein is composed of prolamins and glutelins (Chandna and Matta, 1990).

Albumins contain high levels of lysine and tryptophan whereas globulins are deficient in

these amino acids. Sulphur containing polypeptides were of Mr 47-45 000 which on

reduction yielded polypeptides of Mr 26-25 000.

20

1.6.1.3 Fat

Pearl millet grain has been reported to have a fat content of approximately 5.8% (Osman,

2011). Maize, however, has a lower fat content. Unsaturated fatty acids in pearl millet include

palmitoleic acid (16:1), oleic acid (18:1), linoleic acid (18:2) and linolenic acid (18:3) that

comprise approximately 75% of the total lipid content (Rooney, 1978). The saturated fatty

acids which include palmitic acid (16:0) and stearic acid (18:0) account for approximately

25% of the total fats. The degree of unsaturation of fatty acids contributes to the development

of objectionable odours and flavours after the grinding of pearl millet.

Pearl millet has a higher oil content compared to other cereals because of its relatively large

germ size in relation to the endosperm. Decortication has been found to reduce the fat content

from 5.2 to 4%.

Table 1.1 Proximate and chemical composition (g /100 g) of pearl millet, maize and

cassava

Cereals

Pearl Millet Maize Cassava

Total

Carbohydrates

67.51 62.1-69.94 80.1–86.37

Protein 13.6-15.22,3 8-115 1.2-1.87

Fat 5.83 1.8-4.36 0.1-0.38

Ash

Crude Fibre

2.210

2.810

0.81-1.359

0.79-2.789

1.3–2.87

1.5–3.57

1. Nambiar et al.2011 2. Ali et al., 2003 3. Osman, 2011 4.Demeke.,2018 5. Prasana et al., 2001 6. Nascimento et al.,2014 Ignjatovic-Micic

et al., 2014 7. Charles et al., 2005 8. Montagnac et al., 2009 9. Qamar et al,2016 10. Sawaya et al.,1984

21

The quality of pearl millet flour is improved through the removal of the pearl millet germ

which is rich in unsaturated fatty acids. Consequently, lipid oxidation is prevented, and shelf

life of the flour is extended (Slavin, Jacobs and Marquart, 2001).

1.6.2 Nutritional composition of maize


The starch content of maize ranges between 72 to 73 percent of the kernel weight. Simple

sugars comprise 1 to 3% of the kernel. Maize starch is composed of 21% amylose and 79%

amylopectin (Wilson, Whitt, Ibáñez, Rocheford, Goodman and Buckler, 2004). Amylose

comprises linear chains of α (1→4) linked glucose molecules, whereas amylopectin consists

of a highly branched molecule of α (1→4) linkages with α (1→6) branch points.

Amylopectin promotes the storage of starch in the endosperm of maize kernels. It has a semi

crystalline nature. The functional properties of starch are greatly affected by the amylose-to-

amylopectin ratio (Jane, Chen, Lee, McPherson, Wong, Radosavljevic and Kasemsuwan,

1999). The texture and acceptability of foodstuffs are influenced by the cereal

starch/hydrocolloid mixtures (Alloncle and Doublier, 1991). Gelatinization refers to the

thermal disarrangement of crystalline structures in native starch granules (Tester and

Morrison, 1990). In most food systems the temperature at which pasting behaviour and

rheological properties of the partially or fully swollen starch granules are affected, is more

important than the starch gelatinisation temperature itself. A uniform gel is obtained if the

swollen starch granules are dispersed mechanically as affected by the properties of the starch-

water system.

22

1.6.2.2 Proteins

Zein accounts for approximately 50 to 70% of the total protein found in the endosperm of

maize and is predominantly comprised of the amino acids glutamine, leucine and proline

(Prasanna, Vasal, Kassahun and Singh, 2001). The protein in maize is found mostly in the

endosperm. Maize protein content ranges between 8 to 11 percent of kernel weight. In maize,

globulins located in the germ are high in histidine, arginine, and phenylalanine (Landry and

Moureaux, 1980). Notable variation in histidine, aspartic acid (or asparagine), and proline

have been reported. Proteins in the maize germ have substantial amounts of arginine and

glycine but are deficient glutamic acid (or glutamine). The endosperm contains insoluble

proteins which are rich in histidine and proline but poor in lysine. The Prolamin and G1- and

G2-glutelins extracted from germ and those isolated from the endosperm similarly have a

high proportion of hydrophobic amino acids whereas prolamins are deficient in lysine.

Significant amounts of methionine for G1-glutelins and of histidine for G2-glutelins have

been reported (Tang et al., 2013).

1.6.2.3 Fat

Approximately 90% of the total oil is located in the germ of a mature corn kernel (Jellum,

1971). Maize grain has a fat content that ranges between 1.8% (Nascimento, Mota, Coelho,

Gueifão, Santos, Matos, Gimenez, Lobo, Samman and Castanheira, 2014) and 4.3%

(Ignjatovic-Micic, Vancetovic, Trbovic, Dumanovic, Kostadinovic and Bozinovic, 2014).

Typical unsaturated fatty acids found in maize include 24.1%, oleic acid (18:1), 61.9%

linoleic acid (18:2) and 0.8% linolenic acid (18:3) of the total fatty acids (Weber, 1978). The

saturated fatty acids include 11.1% palmitic acid (16:0) and 2.0% stearic acid (18:0).

23

Research has shown an increase in the fatty acid content in genetically modified corn

varieties (Rayan and Abbott, 2015).

1.6.3 Nutritional composition of Cassava


The major component of cassava grain, as in all cereals, is starch. The starch content may

vary from about 80 to 86% (Charoenkul, Uttapap, Pathipanawat and Takeda, 2011)

comprising both amylose and amylopectin (Charles, Sriroth and Huang, 2005). Total amylose

content has been reported to be 22.4% (Gunaratne and Hoover, 2002). The carbohydrate

content in cassava varies among the genotypes. About 17% sucrose has been found in some

sweet varieties containing low amounts of fructose and dextrose (Charles, Sriroth and Huang,

2005). Starch granule shape has been described as round to variable and 5.0 to 40 µm in size

(Gunaratne and Hoover, 2002). Cassava starches have lower pasting temperatures than other

cereal starches (Waterschoot, Gomand, Fierens and Delcour, 2015). Pasting properties and

gel rigidity are affected by variations in both starch and non-starch components (Charoenkul

et al., 2011). Cassava flour forms pastes at temperatures within a range of 70-74°C.

1.6.3.2 Proteins

Cassava has a relatively low protein content compared with other sources of energy such as

maize and wheat (Ceballosa, Sáncheza, Cháveza, Iglesiasa, Debouck, Mafla, Tohme, 2006).

Cassava flours have been reported to contain crude protein that range between 1.2–1.8%

(Charles, Sriroth and Huang, 2005). Cassava flour amino acid profile (Table 1.2) is similar to

that of other root tubers particularly, essential amino acids like lysine and threonine.

24

Table 1.2 Amino acid content of processed cassava (adapted from Ngudi et al., 2002)1,

pearl millet and maize (adapted from Ejeta et al, 1987)2

Amino

Acids

Cassava (mg/g)1 Dwb Pearl millet (mg/g)2 Maize (mg/ g)2

Asp 0.005 ±0.003 0.088 ±0.001 0.075 ±0.012

Glu 0.013 ±0.000 0.215 ±0.009 0.208 ±0.024

Ser 0.002 ±0.000 0.055 ±0.007 0.047 ±0.003

Gly 0.007 ±0.000 0.035 ±0.004 0.044 ±0.005

His ND 0.024 ±0.000 0.032 ±0.001

Arg

Ala

ND

0.046 ±0.029

0.041 ±0.002

0.087±0.003

0.044 ±0.010

0.076 ±0.012

Thr 0.370 ±0.119 0.043 ±0.001 0.039 ±0.001

Pro 0.039 ±0.002 0.065 ±0.004 0.101 ±0.011

Tyr 0.0147 ±0.002 0.025 ±0.001 0.043 ±0.006

Val 0.025 ±0.000 0.060 ±0.000 0.055 ±0.002

Met ND 0.026 ±0.004 0.022 ±0.002

Cys ND 0.011 ±0.002 0.021 ±0.003

I leu 0.012 ±0.00 0.044 ±0.001 0.038 ±0.001

Leu 0.023 ±0.001 0.112 ±0.004 0.134 ±0.025

Phe 0.015 ±0.000 0.054 ±0.002 0.052 ±0.004

Trp 0.027 ±0.000 ND 0.008 ±0.003

Asp ND

0.088 ±0.001 0.076 ±0.012

Glu 0.006 ±0.000

0.214 ±0.009 0.208 ±0.024

Lys 0.010 ±0.000

0.030 ±0.003

0.029 ±0.006

25

1.6.3.3 Fat

The lipid content of cassava has been reported to range from 0.1% to 0.3% on a wet weight

basis (Montagnac, Davis, and Tanumihardjo, 2009). Typical unsaturated fatty acids found in

cassava include 37.1%, oleic acid (18:1), 14.6% linoleic acid (18:2) and 11.0% linolenic acid

(18:3) of the total fatty acids (Hudson and Ogunsua, 1974). The saturated fatty acids reported

in cassava include 35.1% palmitic acid (16:0) and 2.0% stearic acid (18:0).

1.7 Extrusion

In the food industry extrusion is an important manufacturing process due to its efficiency

(Anton, Fulcher and Arntfield, 2009). Extrusion technology has replaced many outdated

manufacturing processes used in food and feed industry (Lazou and Krokida, 2010).

Extrusion is an intensive operation, which simultaneously combines high temperature,

pressure, and shear force in order to manufacture products with desirable texture properties

(i.e. crispiness, crunchiness) (Meng,Threinen,Hansen, and Driedger, 2010). In extrusion

cooking, food is subjected to unit operations like mixing, cooking, kneading, shearing,

shaping and forming (Stojceska, Ainsworth, Plunkett, Ibanoǧlu, 2009). The basic components

and a schematic diagram of a single screw extruder is illustrated in Figure 1.8. A food

extruder is a device that facilitates the shaping and restructuring of food ingredients (Riaz,

2013). The versatility of extrusion allows for its application in many food processes. The

advantage of using extruders is that it allows for the cooking, forming, mixing, texturizing,

and shaping of food products while ensuring quality retention, high production rates and

reduced costs. The classification of extruders is based on the method of operation (cold or

extruder cookers) and method of construction (single or twin-screw extruders), which operate

on similar principles of operation (Barroca, 2016).

26

Flours fed are subjected to high shear, pressure, and temperature inside the extruder

transform into viscoelastic mass (Patil, Berrios, Tang and Swanson, 2007). Factors such as

feed moisture content and extruder operating parameters affect the degree of transformation.

These material and process parameters govern macromolecular changes that affect the

rheological properties of the dough in the extruder and the extrudate properties

(Meng,Threinen,Hansen, and Driedger, 2010).

Extrusion uses intensive mechanical shear in order to transform raw material into a viscous,

plastic-like dough (Lazou and Krokida, 2010) which is cooked prior to being forced through

the perforation of the die (Figure 1.8).

Figure 1.8. Single screw extruder operation (adapted from Deng et al., 2014)

During extrusion melting of polymers occur resulting in a nucleation of bubbles (Figure 1.9).

The light density of the polymer to “holes” formed by small air bubbles surrounded by the

extruded mass. Due to a moisture flash off process these bubbles grow in size as the melt

27

leaves the extruder die. The expansion is due moisture vaporization due to the high pressure

and the superheated steam. Upon cooling, the viscoelastic matrix becomes glassy causing

expansion and bubble growth to cease. The textural properties of the final extrudate are

influenced by the amount of expansion. Expansion related parameters affect the overall

quality of the extruded product. It is important to analyse the relationships between structure

and texture characteristics in food product development (Lazou and Krokida, 2010).

Figure 1.9 Diagram of extrudate expansion (adapted from Moraru and Kokini, 2003)

1.7.1 Factors affecting the extrusion process

Snack foods of various types have been made from most cereals using extrusion (Lazou and

Krokida, 2010). Second-generation snacks include snacks that are puffed snacks directly

through an extrusion process. Specific quality attributes of snacks would influence the

sensorial perception and acceptance of consumers. In food product development it is

paramount to analyse the relationships between structural and textural characteristics. The

initial composition of feed and extrusion parameters such as feed moisture, feed composition,

28

feed particle size, feed rate, barrel temperature, screw speed, screw configuration, and die

geometry, affect the structural and textural characteristics of extruded snacks (Lazou and

Krokida, 2010; Meng, Threinen, Hansen, and Driedger, 2010). In addition, the protein and

starch digestibility of extrudates are enhanced with extrusion (Arun Kumar, Samuel, Jha and

Sinha, 2015).

Cereals are ideal for manufacturing extruded snacks due to their excellent expansion

properties. However, many are low in nutritional value especially protein and essential amino

acids. Hence, there is an increasing consumer demand for more nutritious snack. Therefore it

was decided to combine cereals with fish to produce nutritionally balanced extruded

products.

1.8 OBJECTIVES OF STUDY

The objectives of the study are:

1. To characterize the nutritional and physico-chemical and rheological properties of

horse mackerel and local cereals such as pearl millet, maize and cassava flours found

in Namibia

2. To identify formulation blends of horse mackerel and pearl millet, maize and cassava

flours in model systems to provide a basis for developing a horse mackerel extruded

snack of high nutritional and eating quality value.

3. To optimize the extrusion process in order to develop an extruded fried snack made of

horse mackerel and cereals with similar functional properties as commercial shrimp

crackers.

4. To determine the functional properties and eating quality of the horse mackerel fish

mince with pearl millet, maize and cassava starch and compare them with commercial

shrimp crackers.

29

5. To determine the influence of raw material quality on the quality of an extruded horse

mackerel snack and total cost.

30

CHAPTER 2

31

CHAPTER 2

2. PROXIMATE AND BIOCHEMICAL COMPOSITION OF THE CAPE HORSE

MACKEREL (TRACHURUS CAPENSIS), PEARL MILLET FLOUR, MAIZE AND

CASSAVA

2.1 Introduction

Cape horse mackerel (Trachurus capensis) is one of the most widely caught pelagic species

in Namibia (Coetzee, Misund and Oechslin, 2001). It belongs to the family Carangidae

(Jardas, Šantić and Pallaoro, 2004). Fish like horse mackerel are an important component of

the dietary protein in Southern Africa (Hara, 2001). The protein content of horse mackerel is

similar to that of round herring (Kűçükgülmez, Celik, Ersoy, and Yanar, 2010) and sardines

(García-Arias, Álvarez Pontes, Garcíá-Linares, García-Fernández, and Sánchez-Muniz,

2003). The variations in chemical composition of horse mackerel may be attributed to

differences in age, genotypes (Cimmaruta, Bondanelli, Ruggi and Nascetti, 2008), nutrition

and seasonal changes as described by (Celik, 2008).

The aim of the investigation was to determine the proximate analyses, and amino acid and

fatty acid profiles of horse mackerel and local cereals found in Namibia.

2.2 Materials and Methods

2.2.1 Materials

Whole Horse mackerel fish and fillets were used in the study. The size of the fish used were

between 100–160 cm and had an average body weight of ±121 g. Whole fish was deboned,

32

cleaned and the fillet separated. Whole horse mackerel fish were purchased from Seawork

Fish Processors (Windhoek, Namibia).Horse mackerel fish fillets (Trachurus capensis) were

provided by Namsov (Pty) Ltd, Walvis Bay, Namibia via air delivery as frozen fish in dry

ice. Fish mince was prepared by blending the fillet with a food processor. Fish mince was

freeze dried and stored at 4°C to be used for analytical tests.

Pearl millet flour (Pennisetum glaucum (L.) R. Br.) (Meme Mahangu brand manufactured by

Namib Mills), maize (Zea mays) flour (Top Score brand, manufactured by Namib Mills,

cassava flour (Manihot esculentus) (Numushe Trading), salt (Cerebos brand) and sugar

(Marathon Sugar brand) of food grade standard were bought from local supermarkets in

Windhoek, Namibia. Flour and extruded samples were kept at 4°C. The fish were sealed in

polyethylene bags and stored at -78°C until analysed.

All reagents were analytical grade and were purchased from Sigma-Aldrich Company Ltd,

Poole, Dorset, United Kingdom.

2.2.2 Methods

The proximate analysis of the samples (fish mince and cereal flour) was carried out according

to the procedures of the Association of Official Analytical Chemists (AOAC, 2016)

33

2.2.2.1 Moisture content

Moisture content of fish was determined by oven drying fish at 100°C for 12 hours (AOAC

Method 950.46B. An empty moisture dish and metal spatula were dried at 100°C for 60

minutes, placed in a desiccator to cool and weighed. Using a metal spatula, horse mackerel

mince (5 g) was weighed into the moisture dish and placed in the oven at 100°C for 12 hours

until a constant weight was obtained. The moisture content of cereals was determined using

the AOAC Method 925.10, oven drying method (AOAC, 2016). Cereal samples were heated

and dried at ±130°C for three hours until a constant weight was obtained. The fish and cereal

samples were transferred into desiccators to cool and weighed. The amount of loss in weight

relates to moisture content and was calculated as:

% Moisture = (W1-W2) x 100

W1 1

where, W1 is the weight (g) before drying the sample and W2 the weight of the sample after

drying.

2.2.2.2 Ash content

Ash in fish mince and cereal flour was determined by incineration in a muffle furnace at

550°C for 24 h, AOAC Method 923.03. Empty silica dishes were placed in the muffle

furnace at 550°C for 24 hours, allowed to cool in the furnace before removing and placed in a

desiccator to cool and weighed. Using a glass rod, horse mackerel mince (7.5 g) and cereal

flour (5 g) was weighed into the pencil labelled silica dishes (in duplicate) and placed on a

hot plate, in a fume cupboard, and charred. The silica dishes with charred samples were

placed in a muffle furnace at 550°C for 24 hours until a white ash was formed.

34

2.2.2.3 Crude fat content

The Soxhlet method, AOAC Method 2003.05 , was used to determine the crude lipid content

in fish mince and cereal flour. The principle of the method is based on the extraction of fat

through the continuous washing and percolation of oil and fat from sample material using an

organic solvent such as petroleum ether and collection of oil in special metal extraction cups.

The solvent was then collected and distilled off and the remaining extract in the metal

extraction cup was dried to evaporate all remaining organic solvent and the remaining lipid

was weighed.

Clean and empty metal extractions cups were placed in the oven at 105°C for 3 hours, placed

in a desiccator to cool and weighed. Horse mackerel mince (5 g) oven dried at 100°C for 12

hours and cereals (5 g) were weighed into graphite pencil labelled thimbles. The labelled

thimbles (Foss, Birchwood, United Kingdom)were attached to a metal ring to hold the

thimbles in the instrument and placed into the instrument to correspond with the numbered

extraction metal cups. Boiling chips were placed in the metal extraction cups and 50 ml of

petroleum ether (60 -80°C) was added and placed into the SoxTec (Foss, Birchwood, United

Kingdom). The thimbles were lowered into the petroleum ether and boiled for 60 minutes at

90°C and rinsed for an additional 20 minutes. After rinsing, the solvent is collected in the

condenser and the extraction cup is placed in an oven to evaporate any remaining solvent.

The thimbles and extraction cups were weighed and the weight of the lipid was calculated by

the difference.

35

2.2.2.4 Fatty acid analysis

Derivation of fatty acid methyl esters (transesterification) in fish and cereal flours.

Transesterification was performed according the method of Saeed and Howell (1999).

Sodium methylate (2 ml) (Sigma–Aldrich Company Ltd, Poole, Dorset, UK) diluted with

methyl tert-butyl ether (MTBE) (Sigma–Aldrich Company Ltd, Poole, Dorset, UK) in 4:6

v/v ratio was added to the oil extract (100 mg) transferred in a 10 ml glass fitted with a screw

cap and mixed by vortexing for 1 minute. Aluminium foil was used to wrap the tube mixture

and allowed to stand for one hour in the dark at room temperature. After one hour, 2 ml of

distilled water and 5 ml chloroform was added and vortex mixed for one minute and

centrifuged at 3000 x g for five minutes to enable phase separation. The upper layer was

removed with a Pasteur pipette and discarded. In order to neutralize the alkali 2 ml of 1 %

citric acid solution was added to the bottom layer. The sample was shaken for 1 minute,

centrifuged for 5 minutes at 3000 x g and the aqueous phase discarded using a Pasteur

pipette. The chloroform extract was evaporated under a stream of nitrogen in a warm water

bath set at 37 °C until a volume of 200 µl was obtained.

FAMEs (1 µl) was injected with a split mode ratio of 100:1 into the Varian gas

chromatograph (Series 3600, Walton-on-Thames, United Kingdom) equipped with an

autosampler and a hydrogen flame ionisation detector. Hydrogen (gas flow 1 ml/min) was

used as carrier gas. The separation was carried out on a SP-2560 silica capillary column (100

m X 0.25 mm ID, 0.20 µm film thickness (Supelco, Bellefonte, PA). The fatty acid methyl

ester (Supelco 37 component FAME Mix) (Sigma–Aldrich Company Ltd, Poole, Dorset, UK)

was used as a standard. The injection port temperature was 250 ºC and the detector

temperature was 260 ºC. Oven temperature was initially 140 ºC held for 5 min and the final

temperature 240 ºC at 4 ºC/min. Fatty acids were identified by comparing sample retention

times with standard retention times. Methyl esters were identified by comparison of the

36

retention times of authentic standards (Supelco 37 component FAME Mix). The fatty acids

were quantified by integration of peak areas and the composition of fatty acids was expressed

as area percentage of total fatty acids. Trans fatty acids were calculated from Standard (Trans

oleic acid C18:1n9t).

2.2.2.5 Protein determination

Crude protein content was determined using the AOAC Method 928.08, by the Kjeldahl

method (with a nitrogen conversion factor of 6.25 for fish and 5.7 for cereal). The sample is

heated with sulphuric acid and a selenium catalyst (5 g K2SO4; 0.15 g CuSO4, 5H2O; 0.15 g

TiO2) resulting in the digestion of all organic material of the sample. Alkaline (sodium

hydroxide) was added to the peptide solution converting nitrogen into ammonia. The

ammonia was distilled into inert boric acid and titrated with dilute acid as described below.

Fish mince and cereal samples (2 g) were placed into a digestion tube containing

concentrated sulphuric acid (20 ml) and two tablets of the catalyst for each sample. The

solution was mixed well and left to stand for 5 minutes. The Kjeldahl tube was placed in a

pre-heated digestion block (450°C) and an extraction tube was attached to remove the acid

vapours. Digestion was for 3 hours after which the tubes were allowed to cool down. Prior to

distillation, distilled water (70 ml) and 20 ml of 40% (w/v) NaOH was added and distilled for

3 minutes. The ammonia formed was collected in a receiver flask containing a few drops of

0.4% methyl red indicator and 30 ml of 4% boric acid, which was dispensed by the

instrument. The solution in the receiver flask was titrated against 0.1 M hydrochloric acid

until colourless. The amount of nitrogen was calculated from the following equation:

37

%N = 100 x (ml of titrant of sample) x 0.0014

Sample weight (g)

The amount of nitrogen was calculated on the basis that 1 ml of 0.1 M HCl = 0.0014 g N.

Crude protein was calculated by the amount of Nitrogen x factor specific for various

products. The factors 6.25, 5.68, 5.87, 5.83, was used for fish, maize, cassava and millets

respectively i.e. total crude protein (%) = Factor x Nitrogen.

2.2.2.6 Determination of amino acids

Total amino acid content of fish and flour was determined according to the method as

described by Bidlingmeyer, Cohen and Tarvin, (1984) and Badii and Howell (2001). This

method involves the hydrolysis of 2.0 g fish mince sample with 6N HCL (15 ml) and 3.0 g

flour samples with 6N HCL (10 ml) at 110°C for 24 h under nitrogen to release free amino

acids. Amino acid standards (Sigma–Aldrich Company Ltd, Poole, Dorset, UK) (20 μl) and

the free amino acids were derivatised with phenylisothiocyanate (PITC) according to the

Waters Pico-Tag method (Badii and Howell, 2001).

2.2.2.7 Derivatisation of amino acids with phenylisothiocyanate PITC

Following the Badii and Howell (2001) method, hydrolysed samples (20 μl) and amino acid

standards were vacuum dried in glass tubes for 20 mins. Then, 20 μl of drying solution (200

μl 1M sodium acetate, 200 μl methanol, and 100 μl triethylamine (TEA) (2:2:1) was added

and vortex mixed and dried under vacuum for 10 mins. Derivitisation solution was prepared

by mixing 20 μl PITC (kept at -20 °C, under nitrogen to prevent degradation), 140 μl

methanol (HPLC grade), 20 μl TEA, and 20 μl Milli-Q water. Derivitisation solution (20 μl)

is added to the residue vortex mixed and left to stand at room temperature for 20 min. The

38

reagent was evaporated under vacuum for 20 mins. The derivatised samples were dissolved in

100 μl of sample buffer (1 L of 1.9% sodium acetate trihydrate + 0.5 ml TEA with pH

adjusted to 6.4 and filtered). To 940 ml of this buffer, 60 ml acetonitrile was added to make

solution B. The hydrolysed fish, flour and amino acids standards were analysed, in triplicate,

by HPLC. High-performance liquid chromatography (HPLC) was used using the Waters

workstation (Waters: Alliance, Waters, UK, Hertfordshire, UK) with an integrated detector

(Dual ʎ absorbance-Waters 2487) and separation module (Waters 2695).

2.2.2.8 Protein extraction

Phosphate buffer of pH 7.5 (50 mM) was prepared by dissolving 1.248 g of sodium di-

hydrogen orthophosphate (Sigma–Aldrich Company Ltd, Poole, Dorset, UK) and sodium

hydrogen orthophosphate 5.96 g in 1 litre of MilliQ water. A second buffer of 0.8 M NaCl

was prepared like phosphate buffer 7.5 with the addition of 46.75 g of NaCl. Both buffers

were adjusted to pH 7.5 using 0.1 N NaOH or 0.1 N HCl. Ten grams of minced fish was

vortex mixed for 1 minute with 30 ml phosphate buffer (7.5). This mixture was centrifuged at

4000 x g for 10 minutes and the supernatant was collected as water soluble proteins and

stored at 4°C. The pellet, to which 30 ml of phosphate buffer containing 0.8 M sodium

chloride was homogenised using an Omnimixer (CAMLAB) homogeniser at 6000 rpm for

three minutes and rinsed off with 20 ml of the 0.8 M NaCl buffer and kept at 4°C for two

hours. Ice was used to surround the extraction cup of the homogeniser in order to prevent the

denaturation of proteins. The homogenate was centrifuged at 4000 rpm for 20 minutes and

the supernatant was collected as salt soluble proteins. The water and soluble proteins were

analysed by Phast gel electrophoresis and protein extractability content using the method

described by (Badii and Howell, 2001).

39

2.2.2.9 Bradford Protein assay

This assay makes use of a ready to use Bradford Reagent (Sigma–Aldrich Company Ltd,

Poole, Dorset, UK). In an acidic medium Coomassie Brilliant Blue G-250 binds to proteins

causing a shift in absorbance within the wavelengths between 465 to 595 nm. The Bradford

Protein Assay method is used to determine the protein content of water soluble and salt

soluble protein extracts. A solution containing 0.1 ml of protein was vortex mixed with

Coomassie protein assay reagent (3 ml) and the absorbance was read at 595 nm within 1 hour

of adding the reagent. A standard curve was prepared from a Bovine Serum Albumin (BSA)

stock solution to give dilutions containing the equivalent of 1.0, 0.75, 0.5,0.25 and 0.125

mg/ml (Bradford, 1976). The mean absorbance (nm) readings of the BSA standard solutions

were plotted on a standard curve. The protein content of an unknown fish sample was

extrapolated from the standard curve and calculated using the following formula:

Concentration(mg/ml) Volume of supernatant (ml)

Protein Concentration = mass of the sample (g)

2.2.2.10 Phast gel electrophoresis

The Badii and Howell (2001) method was used. A staining solution (2% Stock solution) was

prepared by dissolving one tablet of Phast gel blue R in 80 ml of MilliQ water and stirring for

10 minutes. Methanol (120 ml) was added and mixed for three minutes and filtered prior to

use. A destaining solution containing 30% methanol (Sigma–Aldrich Company Ltd, Poole,

Dorset, UK) (150 ml), 10% acetic acid 60% MilliQ water was constituted. A mixture of 10%

glycerol in 10% acetic acid was used as a Preserving solution for 12.5% gel, according to the

method of Hames and Rickwood (1983).

40

Protein samples (100 µl) were mixed with 100 µl of sample buffer (950 µl stock solution and

50 µl mercaptoethanol) in Eppendorf tubes and placed in a boiling water bath for 5 mins and

cooled to 20°C and centrifuged at 10000 rpm for 2 minutes. Samples were stored at 4°C until

analysis.

During the assay, the gel holder was cleaned with MilliQ water and soft tissue paper. At the

end side of the gel holder, 70 µl of MilliQ water was added in a line. The gel envelope was

opened, and the gel was placed carefully on top of the water so as to ensure that no bubbles

occured. The gel’s casing was kept. The buffer holder was left on top of the gel (on either

side) and the cover brought down.

Using a pen, Nescofilm was positioned on the surface of a sample holder. The sample and

standard (2 µl) were added to each hole. The comb was brought down from one side to the

end in order to take the samples. The pre-programmed settings of the instrument were

allowed to proceed. Forceps were used to remove the gel and place the gel face up in the first

row of the staining chamber. Tubes 1,2 and 3 were placed in the order of staining solution

(100 ml), destaining (300) and glycerol solution (100 ml).

2.4 Statistical analyses

All statistical analyses were performed with SPSS version 23 (IBM Corp. Released 2015.

IBM SPSS Statistics for Windows, Armonk, IBM Corp). Mean values of all data were

expressed with standard deviation. Comparisons were made using one-way analysis of

variance (ANOVA).

41

2.5 Results and discussion

The proximate analysis of horse mackerel indicated a high moisture content of approximately

72.0 (g/100 g) (Table 2.1). This is typically associated with fish as moisture content in fish is

influenced by seasonal variation (Bandarra, Batista, Nunes and Empis, 2001). The edible

portion of fish is composed mostly of water; its function is to create the conditions for

biochemical reactions to occur in cells such as the reactivity of proteins. The final moisture

content in the edible portion of fish is influenced by the level of fish freshness, type of

storage methods and the freezing process. The final moisture content of finfish is affected by

drip loss which is the loss of moisture and soluble nutrients. Additionally, there is an inverse

relationship between moisture content and lipid content in pelagic fish such as horse

mackerel.

Table 2.1. Proximate and chemical composition (g /100 g) of whole horse mackerel

(Trachurus capensis)

Parameter Evaluated Fish Species

% Whole Horse Mackerel

Moisture (wet weight basis) 72.0 ±2.1

Ash 3.3 ±0.7

Fat 4.3 ±0.7

Protein (Nx6.25) 16.4 ±2.9

1. Values are means of three determinations ± standard deviation

2. Values are reported on a wet basis

42

The average value of the ash content for horse mackerel was 3.3 g/100g (Table. 2.1). The ash

content was higher than the 1.3% for horse mackerel reported by (Celik, 2008). The variation

observed could have been due to the fact that the whole fish was analysed instead of only the

edible portion. The high mineral content of the head and bones in finfish contributed to the

high ash value observed.

The fat content of the horse mackerel was within the range of 1.4 to 7.5% reported for horse

mackerel (Bandarra, Batista, Nunes and Empis, 2001). The lipid content was about 4.3 g/100

g; this value may vary seasonally. Pelagic fish such as horse mackerel and sardines generally

have a high fat content and are classified as oily or fatty fish. These types of fish contain high

amounts of polyunsaturated fatty acids, which are reported to be beneficial in lowering the

risk of coronary heart disease.

The protein content of horse mackerel (16%) was lower than the 19% crude protein content

reported by Silva, Mendes, Nunes and Empis, (2006) and 21% reported by Badii and Howell

(2006). The difference may be due to the fact that the whole fish which was analysed instead

of only the edible flesh. Nevertheless, horse mackerel is a good source of protein and may be

used to improve the nutritional value of extruded snacks.

Pelagic fish are active and comprise high amounts of proteins and sarcoplasmic proteins

which contain ATPase enzymes (Saaed, 1998). The composition of sarcoplasmic protein

(water soluble proteins) and myofibrillar protein (salt soluble proteins) are shown in Table

2.2. The composition of sarcoplasmic protein (water soluble protein) in horse mackerel was

6.5 mg/ml (Table 2.2) and myofibrillar protein (salt soluble protein) was found to be 7.8

mg/ml. The sarcoplasmic proteins comprise a large group of proteins which include myogens

and enzymes which are soluble in water or solutions with low ionic strength (Hemung and

Chin, 2013). Fish sarcoplasmic proteins constitutes about 25 to 30% of the total proteins. The

43

myofibrillar proteins comprise 40 to 60% of the total proteins found in fish (Sun and Holley,

2010).

Table 2.2. Concentration of myofibrillar and sarcoplasmic proteins in horse mackerel

Type of Protein in

Horse Mackerel sample

Protein concentration (mg/ml in fish

sample) wet basis

Water Soluble Protein (WSP) 6.5 ±0.7

Salt Soluble Protein (SSP) 7.8 ±0.8



44

Standard (2 and 5), Horse mackerel – Salt Soluble Protein (SSP) (3 and 6) Horse mackerel –

Water Soluble Protein (WSP) (4 and 7)

Figure 2.1. Phast gel electrophoresis separation pattern of sarcoplasmic and

myofibrillar protein in horse mackerel

Phast gel-electrophoresis (Figure 2.1) shows the composition of sarcoplasmic proteins,

myofibrillar proteins and reference standards. Lanes 4 and 7 represent sarcoplasmic proteins

with molecular weights of 52 kDa, 20 kDa and 16 kDa. Lane 6 express bands of myofibrillar

proteins with a molecular weight ranging between 17 kDa, actin 45 kDa and myosin 95 kDa

in horse mackerel.

The amino acid composition of horse mackerel is shown in Table 2.3. Horse mackerel had a

high glutamic acid (9.3%) and lysine content (8.9%). The lysine content obtained was similar

to the 9.7% but lower than the 14.2% glutamic acid reported by Erkan et al. (2010).

45

Table 2.3. Amino acid composition of whole horse mackerel

Amino

Acids

Horse Mackerel (mg/g)

Asp 4.5 ±0.02

Glu 9.3 ±0.03

h.pro 0.6 ±0.00

Ser 3.3 ±0.02

Gly 7.7 ±0.03

His 0.8 ±0.27

Arg

Ala

1.0 ±0.00

6.3 ±0.02

Thr 5.8 ±0.02

Pro 5.0 ±0.04

Tyr 1.7 ±0.01

Val 3.7 ±0.02

Met 1.9 ±0.01

Cys 0.1 ±0.01

I leu 2.5 ±0.02

Leu 6.5 ±0.05

Phe 1.1 ±0.07

Trp 0.3 ±0.02

Lys 8.9 ±0.03



46

Glutamic acid is important in cell metabolism (Newsholme et al., 2003). Lysine is an

essential amino acid required in the diet. The fatty acid composition of horse mackerel is

shown in Table 2.4. The values in the table are given as weight percent of product.

Table 2.4. Fatty acid profile of horse mackerel mince

Fatty acid Peak (%) of total

Myristoleic acid C14:0 7.0 ±0.3

Palmitic acid C16:0 28.0 ±1.3

Palmitoleic acid C16:1 7.7 ±0.4

Stearic acid C18:0 12.4 ±0.6

Oleic acid C18:1 31.3 ±1.4

Linoleic acid C18:2 1.8 ±0.1

Linolenic acid C18:3 0.5 ±0.1

Eicosaenoic acid C20:1 4.1 ±0.2

Behemic acid C22:0 0.2 ±0.1

Erucic acid C22:1 5.2 ±0.3

EPA C20:5 1.8 ±0.1

DHA C22:6 0.0 ±0.1

Saturated fatty acids 48.6 ±0.1

Mono-unsaturated

fatty acids

46.9 ±0.1

Poly-unsaturated

fatty acids

4.4 ±0.1


2. Values are means of two determinations ± standard deviation

47

A similar fatty acid profile in Horse mackerel was reported by other researchers (Eymard,

Baron and Jacobsen, 2009).The results show that DHA was absent in the horse mackerel

samples analysed. This is contrary to the 14.84% DHA reported by Saeed and Howell (1999)

in extracted lipids from Atlantic horse mackerel. Horse mackerel contains a high percentage

of palmitic acid (C16:0) and oleic acid (C18:0). The monounsaturated fraction was found to

be mainly oleic acid. Medium fat pelagic fish such as horse mackerel varies significantly in

lipid percentage depending on season, feed intake, spawning and water temperature.

Subsequently this affects the fatty acid profile of the fish. Lipids, however, may store

mercury and polychlorinated biphenyls (PCBs) which are toxic. It is recommended that

vulnerable populations like pregnant women and children limit consumption of fatty fish to

less than twice a week.

The proximate analysis of pearl millet indicated a low moisture content of 12.4% (Table 2.5).

This is typically associated with pearl millet grain that needs to be dried prior to storage.

Drying of cereals is required to prevent the growth of micro-organisms and germination.

Table 2.5. Proximate and chemical composition of pearl millet, maize and cassava flour

Parameter Evaluated Cereal

% Pearl millet Maize Cassava

Moisture 12.4 ±0.5 11.9 ±0.6 7.6 ±0.9

Ash 1.4 ±0.4 1.4 ±0.5 2.6 ±0.4

Protein (Nx5.83) 13.4 ±2.4 5.1 ±2.2 1.6 ±1.1

Fat

Carbohydrates + Crude Fibre

4.6 ±0.6

68.1 ±1.6

1.4 ±0.2

80.1 ±1.8

0.7 ±0.5

87.5 ±1.2



48

Pearl millet was found to have similar ash contents as those observed in maize. Cassava had

the highest ash content among the cereals. The percentage of ash found in pearl millet, maize

and cassava primarily depends on variety, growing location and climate. Ash is the inorganic

mineral residue remaining after all organic matter has been combusted or acid-facilitated

oxidation. In cereals, most minerals are found in the pericarp, aleurone layer and germ. The

removal of these components during milling thus reduces the overall ash content.

Pearl millet had the highest fat content and cassava the lowest. Pearl millet has been reported

to develop a “mousy odour” due to its high crude fat content that undergoes oxidation and

rancidity. Cereals are typically low in fat but high in carbohydrates. The major components of

cereals are starch and protein, whereas the minor components include lipids, non-starch

carbohydrates, phytic acid, vitamins and minerals (Tacer-Caba et al., 2015). Pearl millet has a

sugar content of 2.56% and comprises 0.94%,1.13%, 0.39% and 0.11% sucrose,

glucose/fructose, raffinose and stachyose respectively (FAO, 1995).

Pearl millet had the highest protein content among the cereals. Prolamins and glutelins

comprise 75% of the total protein fraction in pearl millet. High levels of lysine and

tryptophan are present in albumins whereas in globulins these are deficient.

Pearl millet flour had the highest amino acid score among the cereals, while cassava flour had

the lowest (Table 2.6). This may be due to the fermentation step applied in the processing of

cassava flour to reduce some of the antinutrients found in cassava. Cereals are generally low

in lysine and methionine. It is therefore advantageous to add fish. Phenylalanine was the most

abundant amino acid found in the cereals and constituted about 4.7%, 2.7%, and 3.6% of the

crude protein content of pearl millet, maize and cassava flour, respectively. Pearl millet flour

had the lowest (0.1%) lysine score.

49

Table 2.6. Amino acid compositions of Namibian cereal flours

PM – Pearl Millet flour MZ- Maize flour CV - Cassava



Amino

Acids

mg/g

PM MZ CV

Asp 0.9 ±0.01 0.4 ±0.00 0.1 ±0.02

Glu 2.7 ±0.01 3.3 ±0.17 0.2 ±0.01

h.pro 0.08 ±0.00 0.02 ±0.02 0.04 ±0.00

Ser 1.2 ±0.01 1.3 ±0.03 0.1 ±0.00

Gly 1.8 ±0.01 1.4 ±0.04 0.3 ±0.00

Arg

Ala

0.19 ±0.01

2.4 ±0.02

0.31 ±0.05

2.2 ±0.04

0.23 ±0.00

0.3 ±0.01

Thr 1.9 ±0.01 1.3 ±0.13 0.2 ±0.01

Pro 2.4 ±0.01 3.7 ±0.06 0.3 ±0.00

Tyr 0.5 ±0.02 0.6 ±0.01 0.1 ±0.00

Val 1.2 ±0.01 1.0 ±0.02 0.1 ±0.05

Met 0.2 ±0.00 0.3 ±0.01 0.1 ±0.04

Cys 0.1 ±0.00 0.1 ±0.00 0.1 ±0.03

I leu 0.8 ±0.01 0.3±0.00 0.0 ±0.00

Leu 2.7 ±0.02 3.8 ±0.04 0.4 ±0.30

Phe 4.7 ±0.03 2.7 ±0.36 3.6 ±0.03

Trp 1.2 ±0.03 0.3±0.01 0.3 ±0.02

Lys 0.1 ±0.00 0.4 ±0.01 0.5 ±0.05

50

2.6 Conclusion

Characterisation of horse mackerel showed that it has high nutritional value and significance

in its raw form, as fish tend to have a high protein content, a good general amino acid profile

and abundance of polyunsaturated fatty acids. Horse mackerel contained a high percentage of

lipid and protein as shown in the data in this chapter. In particular, Horse mackerel is rich in

omega 3 and omega 6 fatty acids as shown in the fatty acid profile obtained in this study. The

lipid and amino acid profile of horse mackerel indicates that horse mackerel is nutritionally

valuable. The high percentage of protein may be used to solve the problems of protein

shortages around the world, particularly in developing countries. However, it should be noted

that the high percentage of PUFA lipid content makes the fish more susceptible to lipid

oxidation. The following chapters on the extrusion studies were undertaken with a view to

use fish in the extrusion of new food products.

51

CHAPTER 3

52

CHAPTER 3

3. RHEOLOGICAL, THERMODYNAMIC (DSC) AND STRUCTURAL (FT- RAMAN

SPECTROSCOPY) STUDIES OF FISH MINCE, PEARL MILLET, MAIZE FLOUR

AND CASSAVA STARCH RAW MATERIALS

Aim

The aim of this research was to investigate the physico-chemical, structural and rheological

properties of raw materials including horse mackerel and local Namibian cereals including

pearl millet, maize flour and cassava starch. The characterisation of the raw materials and

results obtained by Differential Scanning Calorimetry, FT Raman spectroscopy and small

deformation rheology of the individual raw materials in this chapter formed the basis for

selecting mixtures and formulations to make model products (Chapter 4). The studies from

chapters 3 and 4 led to the final product formulation selected for the extrusion studies and

novel extruded product (Chapter 5).

3.1 Introduction

The nutritional and functional properties of macronutrients are affected by heat processing

and shear during the extrusion process (Stojceska, Ainsworth, Plunkett and İbanoǧul, 2009).

Foods high in protein and starch are subjected to intense mechanical shear resulting in a

viscous dough that passes through the die to form the desired shape (Lazou and Krokida,

2010). Starch and proteins undergo gelatinisation and denaturation respectively in the barrel,

that affects textural properties of the extruded product (Willett, Millard and Jasberg, 1997).

Moreover, the content and type and quantity of protein in the feed may reduce product

53

expansion and lead to a more rigid, hard structure (Chaiyakul, Jangchud, Jangchud,

Wuttijumnong and Winger, 2009). The formation of low density products are reported to

result from substantial denaturation of protein and starch structural break down

(Pansawat,Jangchud, Jangchud, Wuttijumnonga, Saalia, Eitenmiller and Phillips, 2008).

The three-dimensional structure of proteins is altered by heat to expose functional peptide

groups like CO and NH and side chain polar and hydrophobic groups which participate in

hydrogen bonding, electrostatic and hydrophobic interactions (Howell et al., 1999; Guerrero,

Kerry and de la Cabla, 2014). The structure of raw materials may be characterised by FT

Raman spectroscopy, which is a very useful tool to examine changes due to processing.

Protein denaturation is also linked with the breakdown of intramolecular bonds e.g. resulting

in an endothermic process (Frydenberg, Hammershøj, Andersen, Greve, and Wiking, 2016).

Changes in the bonding pattern and the native/denatured state of the protein influence

enthalpy. For example, a reduction in enthalpy is observed in proteins that are in a

conformational state with fewer or weaker bonds as they require less energy to unfold. DSC

was used to monitor the thermodynamic and structural changes which influence the rheology

and texture properties.

This research examined the protein and functional quality of horse mackerel and local cereals

available in the Namibian market.


3.2.1 Small-deformation oscillatory measurements

Measurements of stored elastic energy (G’) and energy loss (G”) due to the viscous properties

of materials including fish mince, either individual flours and mixtures of flours (50 % (w/v)

in distilled water were determined with a modified method of Sarbon, Badii and Howell

54

(2015). The modification was related to the parallel plate geometry gap which was set at 2.0

mm, due to the solid nature of the sample, instead of 0.3 mm (Badii and Howell, 2006).

Samples at the selected concentration in distilled water were placed on the Peltier plate and

after lowering the geometry, the sample was encircled with silicone oil to prevent excess

drying.

3.2.2 Temperature sweeps

The gelation and melting temperature of the individual fish and flour samples were

determined using a dynamic temperature sweep rheological test. Temperatures were ramped

up at intervals between 20°C and 90°C and then cooled down to 20°C on a Rheometrics

Constant Stress 200 rheometer. A constant stress of 0.13 Pa was applied and compared with

the resultant strain. The instrument settings were at 0.1 Pa stress with a frequency of 1 rad/s

and scanning rate of 2°C/min. A holding temperature of 40°C for 10 min on the Peltier plate

allowed equilibrium to be reached. Heating and cooling cycles from 20°C to 90°C and back

to 20°C enables gelation occurring as signified by an increase G’ (elastic modulus). The gel

formation point (Gudmundsson, 2002) occurs at the G’/G” cross over signifying a transition

into a sol-gel. A drop in G’ and an increase in G represents melting which occurs during

reheating from 20°C to 90 °C.

3.2.3 Differential scanning calorimetry (DSC) Measurements

Thermal transition of fish and flour mixtures were determined using a DSC VII calorimeter

(Setaram, Lyon, France). Fish mince was used in its pure form, while flours were mixed with

distilled water to obtain a 50% (w/v) paste. Both fish and flour (600 mg) were placed into

hermetically sealed aluminium sample containers, using water as a reference. Temperatures

55

were raised from 10 °C to 90 °C incrementally at a rate of 0.5°C/min. Heat gained or lost was

expressed as a function of temperature indicating whether an endothermic or exothermic

reaction took place. The area calculated under the curve derived from the onset temperature

(To) and the peak gelatinization temperature (Tm) shown in the thermogram, indicates an

endothermic or exothermic reaction.

3.2.4 FT-Raman spectroscopy

Raman Spectroscopy was carried out following the methods described by (Howell et al.,

1999) and (Badii and Howell, 2003). Fish and flour samples were placed in clear glass vials

and scanned using a Perkin-Elmer System 2000 FT-Raman spectrophotometer. Sulphur,

which forms a peak at 217 cm-1,was used to calibrate the instrument. A beam with a laser

power of 1785 mW was used to analyse samples. The spectra obtained were an average of 64

scans which were baseline-adjusted and standardized according to the Raman shift intensity

of the phenylalanine band obtained at 1004 cm-1. The Grams 32 software (Galactic Industries

Corp., Salem, NH) was used to analyse the recorded spectra. Raman shifts and intensities

were compared with those found in literature (Howell et al.,1999 and Badii and Howell,

2003).


3.3.1 Rheological studies

The results of the viscoelasticity measurements of horse mackerel mince are shown in

Figure 3.1 and detailed in Table 3.1

56

Figure 3.1. A dynamic temperature sweep (20-90-20°C) of horse mackerel mince (100%

w/w)

The gelation profile of the horse mackerel mince (100% w/w) and flours (50% w/w) as a

function of temperature followed different patterns. During heating from 20°C to 90°C, there

was a decrease initially in storage modulus (G’) with a rise in temperature up to 42°C

indicative of myosin unfolding, after which the storage modulus increased considerably and

lowered slightly towards the 90 °C (Figure 3.1). There were slight increases in the loss

modulus (G”) that were not as high as those of the storage modulus (G’). The increase in G’

of the horse mackerel mince during heating is an indication of gel formation. The G’

remained higher than the G’’ as the fish mince was already in a paste or gelled state.

20.0 30.0 40.0 50.0 60.0 70.0 80.0 90.0

103

104

105

Temp [°C]

G

' ()

[P

a]

G

" (

)

[P

a]

100%Fresh Horse Mackerel 1

57

Subsequent cooling from 90 °C to 20 °C resulted in an increase in both the storage and loss

modulus (Table 3.1) that was due to gel setting and an increase in rigidity. This is reported to

be due to the formation hydrophobic interactions and from changes in the orientation and

rearrangement of the polypeptides leading to a stable network during cooling.

The gelation profiles of pearl millet flour, maize flour, cassava flour are shown in Figures

3.2, 3.3 and 3.4 respectively and in detail in Table 3.1.

Figure 3.2 A dynamic temperature sweep (20-90-20°C) of pearl millet flour (50% w/v in

distilled water)

20.0 30.0 40.0 50.0 60.0 70.0 80.0 90.0

104

105

106

Temp [°C]

G

' ()

[P

a]

G

" (

)

[P

a]

100%Pearl Millet flour 3

58

Table 3.1. Storage modulus (G’ Pa) and loss modulus (G” Pa) values for Horse

mackerel mince, pearl millet, maize and cassava flours




4. ND Not determined

Mixture name 20°C before heating 90°C 20°C after heating

G’(Pa) G”(Pa) G’(Pa) G”(Pa) G’(Pa) G”(Pa)

Horse mackerel1

7517

±1878

2130

±401

14706

±1385

1890

±116

56212

±5299

13500

±1698

Pearl millet2

Maize2

Cassava2

55565

±7580

79091

±48676

279330

ND

16409

±3180

27504

±6873

176090

ND

190990

±53712

59469

±15871

124370

±13470

53228

±21363

10735

±1976

37628

±14027

87630

±22054

91911

±11638

1503800

(ND)

13599

±1736

16241

±3519

908330

(ND)

59

The cereals with high amounts of starch content showed an initial decrease in storage

modulus (G’) with a rise in temperature at 45 °C, after which the storage modulus increased

considerably at 66 °C and decreased substantially close to 90 °C (Figures 3.2, 3.3 and 3.4).

Figure 3.3. A dynamic temperature sweep (20-90-20 °C) of maize flour (50% w/v in

distilled water)

After 73°C, there was a marked drop in the G’ and G” values before 90 during heating (Table

3.1). There was no cross-over point as G’ was always higher than the G’’ due to the solid

nature of the flour paste. A dip was observed at a temperature range of 63 to 66 °C indicating

gelatinisation temperatures of cereals. At high temperatures starch granules gelatinised

further resulting is the formation of a matrix of amylose holding the amylopectin in a

network. An increase in the storage modulus is considered to be primarily due to the melting

20.0 30.0 40.0 50.0 60.0 70.0 80.0 90.0

103

104

105

106

Temp [°C]

G

' (

)

[P

a]

G

" (

)

[P

a]

100%Maize flour 2

60

of the crystalline regions of the starch granules and swelling of the starch during

gelatinisation resulting in the formation of a starch granular network.

It is hypothesized that gel formation may occur due to starch and protein interactions.

However, these gels may be unstable as observed by a decrease in storage modulus with

further increase in temperature. The increase in viscosity and gelation during the cooling

period was attributed to the retrogradation of starch, whereby changes in the arrangements of

amylose and amylopectin occur.

Figure 3.4. A dynamic temperature sweep (20-90-20°C) of cassava flour (50% w/v in

distilled water)

20.0 30.0 40.0 50.0 60.0 70.0 80.0 90.0

104

105

106

107

Temp [°C]

G

' ()

[P

a]

G

" (

)

[P

a]

100%Cassava flour 1

61

3.3.2 DSC Studies

The thermograms generated for cereals used in the development of the horse mackerel snack

are shown in Figure 3.5.

Figure 3.5. Typical DSC profiles of cereals pearl millet, maize and cassava (50% w/v in

distilled water)

Table 3.2 shows the calculated enthalpy and transition temperatures for the various cereals.

The peaks observed for pearl millet flour include peak 1 (Tm 69.3 – 72.9 °C). Maize flour

transition: peak 1 was observed at (Tm 63.4 to 67.4 °C). This peak may be due to

gelatinisation. Cassava flour transition appeared at 58.7 to 64.7 °C.

-0.5000

-0.4000

-0.3000

-0.2000

-0.1000

0.0000

0.1000

0.2000

0.00 20.00 40.00 60.00 80.00 100.00

He

at F

low

/Mw

Temperature(°C)

Pearl millet flour

Maize flour

Cassava flour

62

Figure 3.6 Typical DSC profile of horse mackerel mince

The thermogram generated for horse mackerel mince is shown in Figure 3.6. The

thermograms were endothermic with each having three and two transitions respectively.

Table 3.2 shows the calculated enthalpy and transition temperatures for the fish mince and

individual cereals. The transitions observed in the thermograms with maximum transition

temperatures could be assigned to the denaturation of specific constituents of the protein

namely myosin – Peak 1 (Tm, 36.6 °C), water soluble sarcoplasmic proteins Peaks 2 and 3

(Tm 42.4 to 44.7 °C) and lastly actin- peak 4 (Tm 60.17) for the horse mackerel mince.

-0.15

-0.1

-0.05

0

0.05

0.1

0.15

0.00 10.00 20.00 30.00 40.00 50.00 60.00 70.00 80.00 90.00 100.00

He

at F

low

/m

W

Temperature (°C)

63

Table 3.2. Transition temperatures and enthalpy change (ΔH) for pearl millet, maize,

cassava and horse mackerel mince

Sample Peak

Tm °C

Onset Temp

°C

Enthalpy

change J/g

Type of

reaction

Pearl Millet 72.9 69.3 0.3512 Endothermic

Maize 67.4 63.4 0.4494 Endothermic

Cassava 64.7 58.7 3.1039 Endothermic

Horse mackerel

mince

36.6 33.2 0.2423 Endothermic

44.7 42.2 0.0097 Endothermic

63.3 56.8 0.2598 Endothermic


64


The spectra (cm-1) are presented as mean ± standard deviation for relative peak intensity of

the spectral bands (Table 3.3).

Table 3.3 Relative peak intensity of Raman spectra for regions 3200-700 cm-1 for Horse

mackerel mince (freeze dried), pearl millet flour, maize flour and cassava flour

Relative Peak Intensity

Peak assignment

(wavenumber ±

2 cm-1)

Horse

Mackerel

mince

Pearl millet Maize flour Cassava

Trp (760) 0.5a ±0.1 0.5a ±0.0 0.4a ±0.4 0.7b ±0.2

Tyr (830, 0.4a ±0.0 0.4a ±0.1 0.5a ±0.1 0.5a ±0.1

855) 0.4a ±0.2 1.7b ±0.2 1.7b ±0.2 2.2c ±0.4

Helix C-C

stretch,CH3,

symmetric stretch

(937)

0.5a ±0.2 2.3 ±0.4b 2.3b ±0.1 3.0c ±0.3

β-sheet type

structure 990

0.3a ±0.1 0.4a ±0.1 0.4a ±0.0 0.9b ±0.2

Phe,ring band

(1004)

0.6a ±0.3 1.4bc ±0.2 1.3b ±0.1 1.9c ±0.4

Isopropyl 0.4a ±0.2 3.0b ±0.2 3.2b ±0.2 4.4c ±0.6

65

antisymmetric

stretch

CH stretch back

bone (1128)

CH3 anti

symmetric

(aliphatic) CH3

rock (aromatic)

(1160)

0.3a ±0.0 1.3b ±0.2 1.4b ±0.1

2.7c ±0.3

β-sheet type

(1239)

0.6a ±0.2 1.3b ±0.3 0.8a ±0.3 1.0ab ±0.3

Amide III random

coil (1245)

0.7a ±0.1 1.4b ±0.1 1.0ab ±0.3 1.3ab ±0.6

Amide III (1264) 0.8a ±0.1 2.3c ±0.4 1.4ab ±0.4 1.8bc ±0.7

Amide II (1320) 1.2a ±0.2 3.1b ±0.4 1.6a ±0.9 2.3ab ±0.9

H band doublet

from trp (1340)

1.1a ±0.3 4.8b ±0.5 3.1b ±1.3 4.7b ±1.4

(sh*,residue

vibration) asp, glu,

lys (1425)

0.7a ±0.2 2.6b ±0.6 1.0a ±0.9 1.4ab ±1.0

Aliphatic groups,

CH bend (1451)

2.3ab ±0.4 3.5b ±0.4 1.9a ±0.8 3.4b ±1.0

Trp (1554) 0.3ab ±0.1 0.9b ±0.7 0.1a ±0.1 0.4ab ±0.1

Amide I (1660) 1.6b ±0.2 1.4b ±0.7 0.3a ±0.2 0.4a ±0.1

CH stretch, 4.3b ±0.9 4.5b ±0.5 3.0a ±0.3 4.6b ±0.7

66



3. Anova Single Factor was carried out using Duncan’s test at p<0.05

4. Comparison done within rows. Figures with different letters are significantly different

(p<0.05)

The aromatic amino acid phenylalanine showed a strong band at 1007 cm-1 for cassava flour.

It was used in normalizing the spectrum as the intensity of this band is not influenced by the

microenvironment or external factors (Badii and Howell, 2006). Horse mackerel mince had

the lowest intensity values for most spectral bands. The low I855/I830 observed in horse

mackerel mince indicates strong hydrogen bonding (Howell and Li Chan,1996) and correlates

with the lower G’ and G” small deformation values (Table 3.1). Cassava flour had a high

I855/I830 ratio which relates to the tyrosine residue vibrations that have an important role in

hydrogen bond formation highlighting the exposure of its tyrosine residues. In addition,

cassava flour a high vibrational peak of 2.97 at 937-940 cm-1 which relates to a skeletal C-C

stretching and is proportional to the α-helix content.

Furthermore, cassava flour had a high peak intensity of 4.42 in the 1128 cm-1 region which

may be attributed to a higher indole ring vibration of the tryptophan residues in cassava flour.

Cassava flour had the highest peak intensity of 2.73 in the 1160 cm-1 region and this relates to

hydrophobic interactions by intensification of spectral bands assigned to the CH and CH2

bending vibrations and a decrease in the intensity of bands assigned to tryptophan residues in

aliphatic (2940)

Shoulder (2888) 2.1a ±0.3 3.4b ±0.5 2.1a ±0.2 4.5b ±1.1

Shoulder (2976)

(2969)

2.5b ±0.5

2.9ab ±0.5

2.6b ±0.3

2.9ab ±0.4

1.8a ±0.2

2.1a ±0.1

2.8b ±0.4

3.3c ±0.5

67

a nonpolar environment (Howell, Arteaga, Nakai, and Li-Chan, 1999). The high band

intensity of 3.45 observed in cassava flour in the 2969 cm-1 Raman band region relates to the

CH3 asymmetrical stretching at the 2956-2977 cm-1 band. In pearl millet flour, the high

vibrational peak of 2.57 in the 1425 cm-1 relates to the COO symmetric stretch of the ionized

carboxyl group vibration for aspartic and glutamic acid residues (shoulder) and side chain

vibrations of the imidazole ring of histidine (Sarbon, Badii and Howell, 2015). This

corresponds to the higher aspartic acid value of pearl millet flour amino acid profile

compared to maize and cassava flour.

The pearl millet flour had a high relative intensity value of 3.51 in the Amide II Raman band

region of 1320 cm-1 which relates to the interaction between hydrophobic and aliphatic

residues. Sarbon, Badii and Howell (2015) observed C–H deformation (bending and

stretching) of aliphatic amino acids residues in the 1400–1500 cm-1 spectral region. The level

of changes in C–H bending band intensity depend on polarity and micro-environment. These

results concur with the amino acid profile (Table 2.6) which showed higher levels of aromatic

and aliphatic amino acids in pearl millet compared with maize and cassava flour. The high

vibrational peak of 2.32 observed in pearl millet in the spectral and Amide III band region of

1264 cm-1 indicates a significant amount of β-sheet αstructures in pearl millet flour which

may explain their lower G’ and G’ values compared to cassava and maize flour. Moreover, it

was observed that pearl millet showed a high peak value of 0.88 at the Trp 1554 cm-1 Raman

band region. The bands at 760 and 1554 cm-1 that arise from the indole-ring vibrations of the

tryptophan residue (Badii and Howell, 2006). The decrease in the intensity of the tryptophan

band indicates exposure of the hydrophobic residues, thus confirming the role of the

hydrophobic interaction in the low G’ and G” values (Table 3.1).

68

3.4 Conclusion

Rheology is an appropriate method to monitor the changes in fish protein and cereals as

affected by temperature. High temperatures denature protein and gelatinise starch in cereals,

affecting the formation and stability of the gel network that was manifested in the increase in

G’ and G” values. Differential Scanning Calorimetry supported the occurrence of protein

unfolding and denaturation by correlating the transition temperatures with the G’/G” values.

Products of similar textural properties can be developed despite the varying composition and

these properties were applied in the mixtures of fish and cereals in Chapter 4 and extruded

snack product in Chapter 5.

69

CHAPTER 4

70

CHAPTER 4

4. PRODUCT DEVELOPMENT AND PRODUCT MIXTURES

4.1 Introduction

The aim of this research was to investigate the physical and functional properties of

composite flours e.g. cassava and maize with the increasing levels of horse mackerel, pearl

millet flour along with other local Namibian cereals. Cereals are ideal for manufacturing

extruded snacks due to their excellent expansion properties. However, many are low in

nutritional value especially protein and essential amino acids. As there is an increasing

consumer demand for more nutritious snacks, it was decided to combine cereals with fish and

test the physical-chemical and functional properties in model systems with the eventual aim

of producing a nutritionally balanced extruded product in Chapter 5. Results of small

oscillatory deformation tests and Differential Scanning Calorimetry in previous chapters

revealed that heating cereals and fish to high temperatures caused alterations in the behaviour

and interaction between constituent food components mainly proteins and starch. Raman

spectroscopy was used to detect molecular changes associated with gelation and protein

denaturation.

The primary physicochemical properties of starch are its gelatinization and rheological

properties which determine its applications (Yongfeng and Jay‐lin, 2014). During the

gelatinisation of starch, changes occur in its granule crystalline structure which becomes

amorphous. Starch undergoes degradation and its rheological properties are affected (Willett,

Millard and Jasberg, 1997). DSC endotherms of starch are an indication of uptake of water,

71

swelling of amorphous regions and starch crystallites melting (Wang, Zhang, Wang and

Copeland, 2016).

The three-dimensional structure of proteins is altered by heat; this denaturation process

exposes functional peptide groups like CO and NH and side chain polar and hydrophobic

groups which participate in hydrogen bonding, electrostatic and hydrophobic interactions

(Guerrero, Kerry and de la Caba, 2014). Protein denaturation is linked with the breakdown of

intramolecular bonds resulting in an endothermic process (Frydenberg, Hammershøj,

Andersen, Greve and Wiking, 2016). Changes in the bonding pattern may cause changes in

denaturation enthalpies. A reduction in enthalpy occurs in proteins that have a conformational

state with fewer or weaker bonds thus requiring less energy to unfold.

This research examined the protein and functional quality of composite flours made with and

without fish, with local cereals pearl millet, maize and cassava available in the Namibian

market, and composite flours with varying levels of horse mackerel mince and pearl millet

flour. The first mixture, Composite flour (without fish) was a composite consisting of maize

flour, pearl millet flour and cassava flour mixed in a 20:30:30 ratio. The second mixture,

Composite flour (with fish) was a composite flour similar to the first mixture except that

horse mackerel was incorporated at a 20% level. Therefore, the composite flour 2 consisted

of horse mackerel, maize flour, pearl millet flour and cassava flour mixed in a 20:20:30:30

ratio. The physicochemical and functional properties of these model systems are described

below.

72


4.2.1 Determination of amino acids

See section 2.2.2.6 in Chapter 2.

4.2.2 Small-deformation oscillatory measurements

See section 3.2.1 in Chapter 3.

4.2.3 Temperature sweeps

See section in 3.2.2 in Chapter 3

4.2.4 Differential scanning calorimetry (DSC) measurements

See section 3.2.3. in Chapter 3.


See section 3.2.4 in Chapter 3.

73


All statistical analysis were performed with SPSS version 23 (IBM Corp. Released 2015.


expressed with standard deviation. Comparisons were made using the Paired Comparison t-

test for the Raman Spectra.


4.4.1.1 Amino acid analysis of mixture of flours and fish

Table 4.1. shows the % mean value of amino acids found in two flour mixtures. The first

mixture, Composite flour (without fish) was a composite consisting of maize flour, pearl

millet flour and cassava flour mixed in a 20:30:30 ratio. The second mixture, Composite flour

(with fish) was a composite flour similar to the first mixture except that horse mackerel was

added at a 20% level. Thus, the composite flour 2 consisted of horse mackerel, maize flour,

pearl millet flour and cassava flour mixed in a 20:20:30:30 ratio.

In general, flour without fish had a lower amino acid score. Flour with fish added had

significantly (p≤0.05) higher glutamic acid, glycine, alanine, histidine, threonine, proline,

valine, leucine and lysine content. The addition of fish at a 20% level increased the amino

acid levels by 61%. Cereals are generally low in lysine and methionine which are essential

amino acids. It is therefore advantageous to add fish mince to improve the nutritional value of

flour products.

74

Table 4.1. Amino acid compositions of Composite Flour (without fish) and Composite

Flour (with fish)

Amino acids Composite Flour

(without fish)

(mg/g)

Composite Flour

(with fish)

Asp 0.4 ±0.17 1.5 ±0.01

Glu 1.4 ±0.10 4.4 ±0.02

h.pro 0.10 ±0.0 0.96 ±0.0

Ser 0.9 ±0.00 2.1 ±0.01

Gly 1.2 ±0.00 4.5 ±0.13

His 0.3 ±0.00 1.5 ±0.02

Arg

Ala

0.1 ±0.00

1.6 ±0.01

0.4 ±0.01

4.0 ±0.03

Thr 1.2 ±0.01 4.1 ±0.14

Pro 2.1 ±0.00 4.1 ±0.03

Tyr 0.4 ±0.00 1.1 ±0.06

Val 0.7 ±0.01 2.1 ±0.03

Met 0.2 ±0.01 0.8 ±0.01

Cys 0.1 ±0.01 0.1 ±0.05

I leu 0.4 ±0.16 1.5 ±0.01

Leu 2.2 ±0.20 4.5 ±0.02

Phe 3.9 ±0.02 3.5 ±0.02

Trp 0.3 ±0.00 0.1 ±0.00

75

Lys 0.3 ±0.00 4.4 ±0.00



3. Composite Flour (without fish) (25:37.5:37.5 maize:cassava:pearl millet)

4. Composite Flour (with fish) (20:20:30:30 horse mackerel

mince:maize:cassava:pearl millet)

4.4.1.2 Differential scanning calorimetry

Differential scanning calorimetry results including the transition temperatures and the

calculated enthalpy values for the two mixtures with and without fish are illustrated in Table

4.2 . The DSC thermograms for mixture of flours with no fish showed that the first transition

occurred at 64.4 Tm °C. The second transition occurred at 77.2 Tm °C. The enthalpy change

(ΔH) value observed for transitions 1 and 2 was 0.3508 J/g and 0.3591 J/g respectively. The

high transition temperatures indicate that pearl millet flour required more energy for starch

gelatinisation. These peaks may be due to the glass transition, crystallisation, gelatinisation,

melting and retrogradation processes of the mixture of three different flours. The peaks

observed for the flours mixed with fish included peak 1 (Tm 64.1) and Peak 2 75.3°C). This

transition temperature range was slightly lower due to the presence of fish protein. In

addition, (ΔH) values were also lower 0.1770 and 0.2852 for peaks 1 and 2 respectively;

these lower values which indicate interaction between starch and fish proteins. Interestingly,

no peaks were observed at 40 °C that would indicate fish myosin denaturation. This

observation is unusual as the results of the amino acid analyses indicate that Composite flour

(with fish) with fish added had a higher amino acid levels.

76

Table 4.2. Transition temperatures and enthalpy change (ΔH) for Composite flour

(without fish) and Composite flour (with fish)

Sample Peak

Tm °C

Onset Temp

°C

Enthalpy

change (ΔH)

J/g

Type of

reaction

Composite flour 64.4 59.9 0.3508 Endothermic

(without fish) 77.2 72.4 0.3591 Endothermic

Composite Flour

(with fish)

64.1

75.3

60.8

71.1

0.1770

0.2852

Endothermic



3. Composite Flour (with fish) (20:20:30:30 horse mackerel mince:maize:cassava:pearl

millet)

4.4.1.3 Small deformation rheology

The gelation profile of the Composite Flour (without fish) and the Composite Flour (with

fish) as a function of temperature followed different patterns is shown in Figures 4.1 and 4.2

and in detail in Table 4.3. The results of the viscoelasticity measurements of Composite Flour

(without fish) is shown in Figure 4.1.

77

Figure 4.1. A dynamic temperature sweep (20-90-20 °C) of Composite Flour (without

fish) (25:37.5:37.5 maize flour:cassava flour:pearl millet flour) with no horse mackerel

mince

During heating from 20°C to 90°C, the initial reading of storage modulus (G’) occurred at

28°C with a slight rise in temperature up to 33°C after which the storage modulus remained

constant up to 49°C, it increased considerably up to 55°C and 72°C and lowered slightly

towards the 90°C (Figure 4.1). Like the storage modulus, the loss modulus (G”) registered its

first reading at 28°C. There was a slight increase up to 48°C, then it increased considerably

up to 55°C where it remained constant up to 90°C. The G’ values were higher than the G’’

values from the outset indicating that the flour mixtures were already gel –like pastes.

20.0 30.0 40.0 50.0 60.0 70.0 80.0 90.0

10-4

10-3

10-2

10-1

100

101

102

103

104

105

106

Temp [°C]

G

' ()

[P

a]

G

" (

)

[P

a]

Composite flour (without horse mackerel) 3

78

The maintained values in both the storage (G’) and loss (G”) moduli values during the

cooling period (90°C to 20°C) was due to the gel setting and becoming rigid. This is reported

to be due to the formation hydrophobic interactions and from changes in the orientation and

rearrangement of the polymers and forming a stable network during cooling (Howell, 2000).

The visco-elastic behaviour of composite flour with fish is shown in Figure 4.2. During

heating from 20°C to 90°C, there was an initial decrease in storage modulus (G’) with a rise

in temperature up to 40°C, which is reported to be related to the denaturation of fish myosin

(Badii and Howell, 2002) after which the storage modulus increased considerably to 68 °C

and 72 °C and decreased sharply towards 90 °C (Figure 4.2). During cooling, the storage

modulus (G’) increased steadily up to 35 °C, where it decreased slightly to 32 °C before

increasing to 20 °C.

79

Figure 4.2. A dynamic temperature sweep (20-90-20 °C) of Composite Flour (with fish)

(20:20:30:30 horse mackerel:maize:cassava: pearl millet) with horse mackerel mince

Similarly, there was a decrease in loss modulus (G”) with a rise in temperature up to 44 °C

after which the loss modulus increased up to 75 °C then lowered sharply at 90 °C.

Subsequent cooling from 90 °C to 20 °C resulted in a gradual increase. The increase in G’ of

the composite flour during heating is an indication of gelatinisation of the starch and protein

denaturation and gelation in the fish. The constant storage (G’) and loss (G”) moduli values

during the cooling period (90°C to 20°C) is an indication of solid like elastic behaviour

(Yadav, Chhikara, Anand, Sharma, and Singh, 2014).

20.0 30.0 40.0 50.0 60.0 70.0 80.0 90.0

104

105

106

Temp [°C]

G

' ()

[P

a]

G

" (

)

[P

a]

Composite flour (with horse mackerel) 2

80

Table 4.3. Storage modulus (G’Pa) and loss modulus (G”Pa) of Composite Flour

(without fish) and Composite Flour (with fish)




4. Composite Flour (with fish) (20:20:30:30 horse mackerel mince:maize:cassava:pearl

millet)

4.4.1.4 Raman Spectroscopy

The only difference in spectral bands (Table 4.4) between Composite Flour (without fish) and

Composite Flour (with fish) was observed in the aliphatic CH3 groups, which indicates

differences in the hydrophobic groups as indicated by amino acid analysis (Badii and Howell,

2006). This is confirmed by the results of the amino acid levels, which were increased with

the addition of fish. Furthermore, the thermograms for Composite Flour (without fish) and

Mixture name 20°C before

heating

90°C 20°C after heating

G’(Pa) G”(Pa G’(Pa) G”(Pa) G’(Pa) G”(Pa)

Composite Flour (without

fish)

78733

±7164

28888

±2492

98966

±3189

30537

±469

ND ND

Composite Flour (with

fish)

145970

±13743

59442

±7063

103080

±2908

31436

±636

407750

±21503

119510

±30597

81

Composite Flour (with fish) are distinctly similar. It is hypothesized that increasing levels of

starch may limit changes in the protein structural conformations.

Table 4.4. Relative peak intensity of Raman spectra for regions 3200-700 cm-1 for

Composite Flour (without fish) and Composite Flour (with fish)


Peak assignment (wavenumber ± 2 cm-1) Composite Flour

(without fish)

Composite

Flour

(with fish)

Trp (760) 0.45a ±0.0 0.46a ±0.1

Tyr (830,Composite 0.42a ±0.2 0.33a ±0.1

855) 2.29a ±0.4 1.63a ±0.3

Helix C-Cstretch,CH3,symmetric stretch (937) 2.83a ±0.1 2.14b ±0.1

β-sheet type structure (990) 0.34a ±0.0 0.25a ±0.1

Phe,ring band (1004) 1.59a ±0.1 1.42a ±0.1

Isopropyl antisymmetric stretch 4.26a ±0.7 3.49a ±0.4

CH stretch back bone (1128)

CH3 antisymmetric (aliphatic) CH3 rock

(aromatic) (1160)

1.60a ±0.1 1.45a ±0.1

82




4. Composite Flour (with fish) (20:20:30:30 horse mackerel mince:maize:cassava:pearl millet)

5. Data from an Independent t- Test significance at 0.05

β-sheet type (1239) 0.50a ±0.2 0.75a ±0.2

Amide III random coil (1245) 0.68a ±0.1 1.06a ±0.3

Amide III (1264) 1.26a ±0.1 1.55a ±0.4

Amide II (1320) 1.48a ±0.2 1.70a ±0.9

H band doublet from trp (1340) 3.48a ±0.4 3.55a ±1.0

(sh*,residue vibration) asp, glu, lys (1425) 0.77a ±0.8 1.14a ±0.8

Aliphatic groups, CH bend (1451) 1.89a ±0.6 2.18a ±0.5

Trp (1554) 0.12a ±0.2 0.19a ±0.1

Amide I (1660) 0.28a ±0.1 0.45a ±0.1

CH stretch, aliphatic (2940) 4.79a ±0.4 4.55a ±0.6

Shoulder (2888) 3.48a ±0.4 3.19a ±0.5

Shoulder (2976)

(2969)

2.71a±0.2

3.25a ±0.3

2.61a ±0.4

3.02a ±0.5

83

4.4.2 Cereal (maize and cassava) composite flours with varying proportions of horse

mackerel mince and pearl millet flour (Composite flour 1, 2, and 3)

4.4.2.1 Differential Scanning Colorimetry of Composite flour 1,2 and 3

The composition of the three composite flours with varying proportions of horse mackerel

and pearl millet flour are presented in Table 4.5 and the transition temperatures for three

composite flours are presented in Table 4.6.

Table 4.5 Composition of composite flours

Composite flour Composition (%w/w)

Horse

mackerel

Maize Cassava Pearl millet

Composite flour 1 10 20 30 40



The thermogram generated for horse mackerel mince (10%), maize flour (20%), cassava flour

(30%) and pearl millet flour (40%), i.e. Composite flour 1 (CF1) indicated that the first

transition occurred at Tm 63.4 °C. The second transition occurred at Tm 76.7 °C. The

enthalpy change (ΔH) value observed was 0.626 J/g and 0.780 J/g for the first and second

transitions respectively. This enthalpy change value is the highest among the three composite

flours. The high transition temperatures agree with the findings of this research that the

composite flour containing higher levels of pearl millet requires more energy for starch

gelatinisation.

84

The peaks observed for Composite flour 2 (CF2) containing 20% horse mackerel, 30% pearl

millet, 20% maize flour and 30% cassava flour include Peak 1- (40.2 - 43.0°C), Peak 2 (Tm

61.0 to 65.3°C) and Peak 3 (Tm 74.1 to 77.8°C). These peaks may be due various stages

which starch undergoes during gelatinisation.

For Composite flour 3 consisting of 30% horse mackerel, 20% maize flour, 30% cassava

flour and 20% pearl millet flour, the transitions appeared at Peak 1- (40.0 - 42.0°C), Peak 2

(Tm 61.0 to 64.7°C) and Peak 3 (Tm 76.5 to 77.5°C), which are similar transition

temperatures to CF2. The lower enthalpies change (ΔH) values observed in CF3 suggests that

less energy was required for protein denaturation which is expected due to an increase of

protein through the addition of fish. Higher transition temperatures were required in

composite flours with more starch.

85

Table 4.6. Transition temperatures and enthalpy change (ΔH) for composite flours with

varying proportions of horse mackerel mince (10%, 20% or 30%) and pearl millet flour

(40%, 30% or 20%)

Sample Peak

Tm °C

Onset Temp

°C

Enthalpy

change J/g

Type of

reaction

Composite flour 1 63.4 58.6 0.626 Endothermic

76.7 72.3 0.780 Endothermic

Composite flour 2 43.0 40.2 0.128 Endothermic

65.3 61.0 0.540 Endothermic

Composite flour 3

77.8

42.0

64.7

77.5

74.1

40.0

61.0

76.5

0.194

0.063

0.397

0.390

Endothermic

Endothermic

Endothermic

Endothermic

1. Values are reported on a wet basis. Each Composite flour was made up to 1:1 with distilled water

3. Composite flour 1 (10:20:30:40 horse mackerel mince:maize:cassava:pearl millet)

4.Composite flour 2 (20:20:30:30 horse mackerel mince:maize:cassava:pearl millet)


86

4.4.2.2 Small deformation rheology

Figure 4.3 A dynamic temperature sweep (20-90-20°C) of Composite flour 1

(10:20:30:40 horse mackerel mince:maize:cassava:pearl millet)

The visco-elastic behaviour of Composite flour 1 is shown in Figure 4.3 and in detail in Table

4.7. During heating from 20°C to 90°C, there was an initial decrease in storage modulus (G’)

up to 49°C with a rise in temperature up to 68°C after which the storage modulus decreased

slightly to 75°C remaining constant towards 90°C (Figure 4.3). During cooling, the storage

modulus (G’) increased slightly down to 69°C, and then increased from 69 to 20°C.

Similarly, there was a decrease in loss modulus (G”) with a rise in temperature up to 46°C

after which the loss modulus values increased to 69°C then lowered gradually to 90°C.

20.0 30.0 40.0 50.0 60.0 70.0 80.0 90.0

103

104

105

106

Temp [°C]

G

' ()

[P

a]

G

" (

)

[P

a]

10%HorseMackerel20%Maize30%Cassava40%PearlMillet 3

87

Subsequent cooling from 90°C to 20°C resulted in a gradual decrease in loss modulus until

75°C followed a constant drop to 20°C.

Figure 4.4. A dynamic temperature sweep (20-90-20°C) of Composite flour 2


The composite flour with a high amount of pearl millet flour and low horse mackerel resulted

in the formation of a weak gel. The constant storage (G’) and loss (G”) during the cooling

period (90°C to 20°C) is an indication of solid like elastic behaviour (Yadav, Chhikara,

Anand, Sharma, and Singh, 2014).

The gelation profile of Composite flour 2 is shown in Figure 4.4 and in detail in Table 4.7.

During heating from 20 °C to 90 °C, there was an initial decrease in storage modulus (G’)

with a rise in temperature up to 69 °C after which the storage modulus decreased gradually

20.0 30.0 40.0 50.0 60.0 70.0 80.0 90.0

103

104

105

Temp [°C]

G

' ()

[P

a]

G

" (

)

[P

a]

20%HorseMackerel20%Maize30%Cassava30%Pearl Millet 3

88

the 90 °C (Figure 4.4). During cooling, the storage modulus (G’) slightly increased up to 63

°C, gradually decreasing to 20 °C. Similarly, there was a decrease in loss modulus (G”) with

a rise in temperature up to 55 °C after which the loss modulus increased to 68 °C then

lowering sharply to 90 °C. Subsequent cooling from 90 °C to 20 °C resulted in a sharp

increase in loss modulus until 69 °C followed by a sharp increase at 69 °C and a gradual

decrease to 44 °C.

Figure 4.5. A dynamic temperature sweep (20-90-20°C) of Composite flour 3


There was a gradual increase in values from 44 °C to 20 °C. The increase in G’ of the maize

flour during heating is an indication of gelatinisation. The constant storage (G’) and loss (G”)

during the cooling period (90°C to 20°C) exhibits a strong viscoelastic gel-like behaviour,

which is typical for of flours with a higher protein content.

20.0 30.0 40.0 50.0 60.0 70.0 80.0 90.0

103

104

105

Temp [°C]

G

' ()

[P

a]

G

" (

)

[P

a]

30%HorseMackerel20%Maize30%Cassava20%PearlMillet 3

89

The gelation profile of Composite flour 3 is shown in Figure 4.5 and in detail in Table 4.7.

During heating from 20 °C to 90 °C, there was an initial decrease in storage modulus (G’)

with a rise in temperature up to 44 °C after which the storage modulus increased sharply to

69 °C then decreased to 90 °C (Figure 4.5). During cooling, the storage modulus (G’)

increased constantly until 20 °C. Similarly, there was a decrease in loss modulus (G”) with a

rise in temperature up to 46 °C, after which the loss modulus increased to 69 °C then lowered

sharply to 90 °C.

Table 4.7. Storage modulus (G’ Pa) and loss modulus (G” Pa) values of composite flours

with varying proportions of horse mackerel mince (10%, 20%, 30%) and pearl millet

flour (40%, 30%, 20%)

Mixture name 20°C

before

heating

90°C 20°C after

heating

G’(Pa) G”(Pa G’(Pa) G”(Pa) G’(Pa) G”(Pa)

Composite flour 1

17607

±8288

5990

±2334

58597

±3808

15119

±1966

132790

±42085

15333

±7477

Composite flour 2

84761

±11914

25801

±3870

52885

±2653

11410

±298

84283

±14443

19608

±822

90


2. Values are reported on a wet basis. Each Composite flour was made up to 1:1 with distilled water




Subsequent cooling from 90 °C to 20 °C resulted in a decrease of loss modulus (G”) until 88

°C where there was a gradual increase in loss modulus until 63°C. This was followed by a

lowering at 55°C and a remained constant to 20°C. The increase in G’ of Composite flour 3

during heating is an indication of gelatinisation.

4.5 Conclusion

The high nutritional significance of horse mackerel in its raw form is more evident when it is

added to cereal flours. The fish tend to have a high protein content, a good general amino

acid profile thereby improving the nutritional composition of composite flours. Despite the

high fish content, Composite flour 3 had the lowest G’ values, probably because pearl millet

content was the lowest and this may be important for gel strength. Composite flour 2

consisting of a 20:20:30:30 horse mackerel, maize flour, pearl millet flour and cassava flour

respectively showed the best starch and protein interactions as evidenced by thermal analyses

and rheological properties. Therefore this formulation was used in the development of the

horse mackerel extruded snack in Chapter 5.

Composite flour 3

21418

±704

5932

±187

38597

±496

8873

±329

47559

±6878

9712

±1918

91

CHAPTER 5

92

CHAPTER 5

5. NUTRITIONAL AND BIOCHEMICAL PROPERTIES OF EXTRUDED CAPE

HORSE MACKEREL (TRACHURUS CAPENSIS), PEARL MILLET FLOUR, MAIZE

AND CASSAVA PRODUCTS

Aim

The aim of the investigation was to develop a fish-based extruded snack product made from

local cereals and Cape Horse Mackerel found in Namibia. Using a low-cost single screw food

extruder (Figure 5.1) built for developing countries, fish based extruded snacks were

developed by incorporating horse mackerel, maize, cassava and pearl millet flour of ratios of

10:20:30:40; 20:20:30:30 and 30:20:30:20 respectively, using barrel temperatures of ambient,

90°C and 120°C (Figure 5.2). Biochemical properties of the extruded snack product were

assessed and the high nutritional quality was compared to popular crackers.

5.1 Introduction

Namibia’s undernourished persons make up approximately 34% of the total population

(FAO, 2012). This is a serious nutritional problem which can be addressed through the

utilization of local and rich sources of protein such as horse mackerel that is abundant in

Namibia.

Most juvenile horse mackerel are considered low value fish and are used primarily for

fishmeal production. This underutilization is due to the non-availability of technology which

93

can transform horse mackerel into shelf stable and acceptable food products. A small

proportion of fish that is used for human consumption is mainly sold fresh, frozen, salted,

dried or smoked or a combination of these processes. By creating a variety of foods made

from horse mackerel, its utilization may be enhanced.

Traditionally, fish consumption has been low in Namibia, with most people preferring meat.

The Namibian government has made concerted efforts to increase local fish consumption.

Therefore, the development of alternative fish products may encourage increased

consumption. Paramount to this study is therefore the development of a nutritious snack food

made from horse mackerel using extrusion technology. Extrusion enables the development of

food products with different textures and shapes (Cheng and Friis, 2010) by using a single

process of cooking using high heat and pressure. In this study the extruded product was

produced from horse mackerel and composite flour made from locally available cereals.

Emphasis was placed on the optimisation of the extrusion process (Pansawat, Jangchud,

Jangchud, Wuttijumnong, Saalia, Eitenmiller and Phillips, 2008) to develop a snack food

from horse mackerel. Specifically, the study investigated the development of a semi-finished

starch containing extruded snack similar to the commercial shrimp crackers available in the

market in Namibia.

Extrusion has the potential of denaturing and modifying the protein (Chaiyakul, Jangchud,

Anuvat Jangchud, Wuttijumnong, Ray, 2009) and gelatinising starch to produce desirable

functional properties (Chaudhary, Miler, Torley, Sopade and Halley, 2008) in the extruded

snack product. This increases the opportunity for the utilization of low value horse mackerel

for human consumption and the diversification of horse mackerel products in the fish

industry. Characteristics such as physical qualities, chemical composition and texture profiles

influence the overall quality of the extruded snack products.

94

This research examined the effects of extrusion processing on the protein and functional

quality of an extruded snack food derived from horse mackerel.


5.2.1 Materials

Whole Horse mackerel fish deboned and cleaned were used in the study. Whole horse

mackerel fish was purchased from Seawork Fish Processors (Windhoek, Namibia). Pearl

millet flour (Pennisetum glaucum (L.) R. Br.), Maize (Zea mays) flour, cassava flour

(fermented) (Manihot esculentus), salt and sugar of food grade standard were bought from a

local supermarket in Windhoek, Namibia. Flour and extruded samples were kept at 4°C. The

fish were held at -20°C until used.

5.2.2 Methods

5.2.2.1 Laboratory-scale single-screw extruder

A single-screw extruder (Figure 5.1) with a 50 mm barrel diameter and 700 mm barrel length

(i.e. length-to-diameter ratio of 14) was used in this research. The barrel was heated by

quartz-based heating bands around the feed, compression and metering sections. The length

of the die was 15 mm and the diameter was 6 mm. The screw clearance had a gap of 2 mm. A

commercial dough roller was used to flatten the mass exiting the extruder.

95

Figure 5.1. Single screw extruder used to process a horse mackerel extruded snack

(University of Namibia)

96

Mincing horse mackerel fish

Cooking the horse mackerel mince in a microwave oven (1500 W) for 7 min

Chopping finely in a food blender

Adding 1 N NaCl (2%), Sucrose (1%) and blend mixing in a blender for 2 mins

Blending Horse mackerel mince, maize flour, cassava flour and millet flour mixed in the

proportion 10:20:30:40 or 20:20:30:30 or 30:20:30:20 (w/w)

Extruding (single screw) each of three well mixed fish-pearl millet, maize, cassava starch

dough blends (Zone 1 – 30°C, Zone 2 – 90°C, Zone 3 – 120°C)

Rolling with a dough roller

Drying overnight at 55°C until the rolled extrudate has a moisture content 10% (w/w)

Slicing the non-expanded snack

Storing each of three products in vacuum sealed plastic containers at 4 oC

Frying the extruded fish cracker in sunflower oil when required

Figure 5.2. Flow diagram of the development of a horse mackerel extruded snack

97

The horse mackerel extruded snack samples were processed at the Department of Food

Science and Technology, University of Namibia, Windhoek, Namibia. Samples were vacuum

sealed in polyvinyl chloride bags and transported by air over a three-day period. These

samples were stored at 4°C and analysed at the Department of Nutrition and Metabolism,

Faculty of Health and Medical Sciences, University of Surrey.

5.2.2.2 Moisture content


5.2.2.3 Protein determination


5.2.2.4 Fat content


5.2.2.5 Ash content


98

5.3 Biochemical characterization

5.3.1 Fatty acid analysis


5.3.2 Amino acid analyses

5.3.2.1 Determination of amino acids


5.3.2.2 Derivatisation of amino acids with phenylisothiocyanate (PITC)


5.3.3 Die swell

Die swell was obtained by measuring the thickness of the extruded mass using a Vernier

calliper in comparison to the aperture size of a 6 mm die.

5.3.4 Expansion Ratio

The expansion ratio (thickness) was determined by measuring the thickness of the disc with a

digital Vernier calliper before and after frying. The mean thickness of 20 samples was taken.

99

5.3.5 Texture analyses

The degree of hardness of fried extruded fish crackers was measured using a Stable Micro

System TAX T2 (Stable Microsystems Ltd., Godalming, UK) attached to a three-point bend

rig (HDP/3PB). The supporting blades of the bend rig were placed 20 mm apart. The samples

were placed in the centre of the support. The Texture Analyser was calibrated with 50 kg load

cell and then followed by the calibration of the probe. The setting mode measured

compression and holding time. The probe compressed the sample until it reached 50% of the

product height, before returning to it starting position.

The level of hardness of the uncooked and fried extruded fish and shrimp crackers was

calculated from the area under the force-time curve using the Texture Exponent software.

The mean of ten randomly selected samples was calculated. The parameters used were a

pretest speed of 2 mm/s, test speed of 0.5 mm/s, post-test speed of 10.00 mm, distance of 5

mm and load cell of 50 kg.


All statistical analysis were performed with SPSS version 23 (IBM Corp. Released 2015.


reported with a standard deviation. Comparisons were made using one-way analysis of

variance (ANOVA).


Composite flour 1 had the lowest die swell value (Table 5.1) compared to Composite flour 2

which had the highest. The low swelling of the extrudate in Composite flour 1 is due to the

100

low fish content in that mixture. Composite flour 1 produced an extrudate that was brittle due

to the uncooked starch.

Table 5.1. Die swell and expansion (%) of the three product composite flours (before

and after extrusion compared to a commercial shrimp cracker.






Measurement Composite

flour 1

Composite

flour 2

Composite flour

3

Shrimp

Cracker

(Control)

Die swell (mm) 7.1 ±0.4 7.7 ±0.1 7.2 ±0.6

Before Rolling

(mm)

7.2 ±0.5 7.3 ±0.5 7.8 ±0.3

After rolling

before drying

(mm)

3.0 ±0.2 4.2 ±0.1 3.0 ±0.4

After drying

before Frying

(mm)

2.8 ±0.3 3.5 ±0.4 2.8 ±0.2 1.3 ±0.3

After frying

(mm)

4.5 ±0.2 6.1 ±0.5 5.6 ±0.4 3.9 ±0.3

Expansion (mm) 1.7 ±0.4 2.6 ±0.3 2.7 ±0.4 2.5 ±0.3

Expansion% 62.1 73.2 97.2 188.8

101

An increase in fish content resulted in firmer extrudates as observed in Composite flour 2 and

Composite flour 3 (Figure 5.3). Furthermore, Composite flour 1 had the lowest percent

expansion of 62.1% compared to 97.2% observed in Composite flour 3. Commercial shrimp

cracker expansion was almost double that observed for Composite flour 3. Although,

Composite 3 had the highest expansion percentage, Composite flour 2 consisting of

20:20:30:30, maize, cassava, pearl millet and horse mackerel mince was considered as the

best product mixture that would result in the optimum quality extruded fish snack (Table 5.2).

Table 5.2. Proximate and chemical composition (g /100 g) of uncooked Horse mackerel

extruded snack and Fried Commercial shrimp cracker

Parameter Evaluated

%

Uncooked Horse

mackerel extruded

snack

Fried Commercial Shrimp cracker

(Adapted from Omobuwajo,2003)

Moisture 21.7 ±1.5 1.9 ±0.18

Ash 4.6 ±0.1 0.7 ±0.18

Fat 2.7 ±0.2 32.2 ±1.31

Protein (N x 6.25) 19.9 ±1.0 2.6 ±0.13



Table 5.3. shows the mean value of amino acids (%) found in horse mackerel extruded

product and a commercial shrimp cracker.

102

Table 5.3. Amino acid composition of uncooked horse mackerel extruded snack

(UHME) and commercial shrimp cracker (CSC)

Amino acids

(mg/g)

UHME CSC

Asp 0.7 ±0.0

Glu 2.3 ±0.0 0.3 ±0.2

h.pro

Ser 1.2 ±0.0

Gly 2.6 ±0.0 0.3 ±0.3

His

Arg

Ala

0.2 ±0.0

2.2 ±0.0

Thr 2.2 ±0.0

Pro 2.7 ±0.0

Tyr 0.6 ±0.0

Val 1.0 ±0.0 0.2 ±0.0

Met 0.4 ±0.0

Cys 0.3 ±0.3

I leu 1.3 ±1.0

Leu 1.7 ±1.4 0.2 ±0.0

Phe 3.5 ±0.0 2.1 ±1.3

Trp 0.0 ±0.0

Lys 1.7 ±0.0

103


2. UHME (20:20:30:30 horse mackerel, maize, cassava, pearl millet flour)

This is because after frying, despite its higher expansion percentage, Composite 3 extrudate

produced a product (Figure 5.4) that had a strong fish smell. In addition, it would not be

practical for snack industry to produce an extruded product with 40% fish as this would be

too cost ineffective.

104

b

)

a)

Figure 5.3 a) Uncooked Commercial Shrimp Cracker b) Uncooked Horse mackerel

non-expanded extruded snack Composite flour 1 10:20:30:40 horse mackerel

mince:maize:cassava:pearl millet) c) Uncooked Horse mackerel non-expanded

extruded snack Composite flour 2 20:20:30:30 horse mackerel

mince:maize:cassava:pearl millet) d) Uncooked Horse mackerel non-expanded

extruded snack Composite flour 3 30:20:30:20 horse mackerel


c) d)

105

Figure 5.4 a) Fried Commercial Shrimp Cracker b) Fried Horse mackerel non-

expanded extruded snack Composite flour 1 (10:20:30:40 horse mackerel

mince:maize:cassava:pearl millet) c) Fried Horse mackerel non-expanded extruded

snack Composite flour 2 (20:20:30:30 horse mackerel mince:maize:cassava:pearl millet)

d) Fried Horse mackerel non-expanded extruded snack Composite flour 3 (30:20:30:20

horse mackerel mince:maize:cassava:pearl millet)

a) b)

c) d)

106

The Commercial Shrimp Cracker generally had a low amino acid score compared to the

uncooked horse mackerel extruded expanded snack (Table 5.3). It is therefore advantageous

to add fish that is high in protein.

In Chapter 4, it was observed that the nutritional value of composite improved with the

addition of fish mince. However, after extrusion most amino acid levels decreased by

approximately 50% which may be due to denaturation of proteins from the high temperatures

used during extrusion. Examples of heat sensitive amino acids include proline, arginine,

histidine, cysteine, threonine, lysine, tryptophan, serine, and methionine (Weder and Sohn,

1983). This decrease in amino acid levels may appear as a significant loss. However,

considering that the flours without fish had 17.8% amino acid level, an uncooked horse

mackerel extruded snack with 24.6% amino acid level is more nutritious any extruded

product made from only cereal flours.

The uncooked horse mackerel extruded product had a better amino acid profile compared to

the commercial shrimp cracker. Commercial shrimp cracker is composed mainly of cassava

starch. Its poor amino acid content maybe therefore be also attributed to the low amino

content found in cassava.

Table 5.4 shows the fatty acid profile of a fried horse mackerel extruded snack made from

Composite flour 2 (20:20:30:30 horse mackerel mince:maize:cassava:pearl millet)

comprising 1.4% saturated (SFA), 2.6% monounsaturated (MUFAs) and 4.3%

polyunsaturated acids (PUFAs). Among them, those occurring in high proportions were

myristic acid (C14:0, 0.05%), palmitic acid (C16:0, 0.92%), palmitoleic acid (C16:1, 0.05%),

107

stearic acid (C18:0, 0.34%), oleic acid (C18:1n9 cis, 2.48%), linoleic acid (C18:2n6, 4.12%),

octadecatetraenoic acid (C18:3n3, 0.105%), C20 (0,03%),C20:5n3, 0.04%) and (C22:0),

0.06%).

Table 5.4. Fatty acid profile of the fried horse mackerel extruded expanded snack

(FHMEES)

Fatty Acid (w/w)

C14 0.1

C16 0.9

C16:1ɷ7 0.1

C18:0 0.3

C18:1 ɷ9 2.5

C18:2 ɷ6 4.1

C18:3 ɷ3 0.1

C20 0.0

C20:1 ɷ9 0.0

C20:5 ɷ3 0.0

C22:0 0.1

Total saturated 1.4

Total monounsaturated 2.6

Total polyunsaturated 4.3

Total ɷ3 content 0.1


108

P/S ratio 3.1

Total Fat 8.3


2. FHMEES (Made from Composite flour 2 (20:20:30:30 horse mackerel


Most of the fatty acids present in the fried extruded horse mackerel snack are from the oil

that it was fried in and absorbed.

Table 5.5. Fatty acid profile of sunflower oil (adapted from (Orsavova et al., 2015)

Fatty Acid (w/w)

C12 0.02

C14 0.09

C16 6.2

C17:0 0.02

C18:0 2.8

C20 0.21

C16:1 (n-7) 0.12

C18:1 cis (n-9) 28.0

C20:1 (n-9) 0.18

C18:2cis (n-6) 62.2

C18:3 (n-3) 0.16

Total saturated 9.4

109

Total monounsaturated 28.3

Total polyunsaturated 62.4



Most of the health beneficial fatty acids are significantly reduced due the incorporation of

cereals with fish in the product mixture. These may also be lost due to the high temperatures

during extrusion and frying.

Figure 5.5. A dynamic temperature sweep (20-90-20°C) of an uncooked Horse mackerel

non-expanded extruded snack made from Composite flour 2 (20:20:30:30 horse

mackerel mince:maize:cassava:pearl millet). Flour mixed with distilled water in 1:1

ratio.

110

In terms of the rheological tests, during heating from 20 °C to 90 °C, there was a decrease in

storage modulus (G’) up to 70 °C after which the storage modulus remained constant until 90

°C (Figure 5.5). Subsequent cooling from 90 °C to 20 °C resulted in a constant storage

modulus. This is due to the setting of gel and an increase in rigidity of the paste made from

the uncooked Horse mackerel non-expanded extruded snack during cooling.

Similarly, there was a decrease in loss modulus (G”) up to 78 °C which then remained

constant to 90 °C. Subsequent cooling from 90 °C to 20 °C resulted in a constant loss

modulus until 60 °C followed by a gradual increase from 60 °C to 25 °C. G’ was always

higher than the G’’ due to the solid nature of the uncooked Horse mackerel non-expanded

extruded snack.

For the commercial shrimp cracker, during heating from 20 °C to 90 °C, there was a sharp

increase in storage modulus (G’) with a rise in temperature up to 47 °C after which the

storage modulus decreased considerably to 86 °C and increased towards 90 °C (Figure 5.6).

During cooling, the storage modulus (G’) increased sharply up to 78 °C, where it remained

constant until 75 °C then increasing sharply to 20 °C. In contrast, the loss modulus (G”)

remained constant up to 45 °C after which the loss modulus decreased sharply to 80 °C then

increased significantly to 90 °C. Subsequent cooling from 90 °C to 20 °C resulted in a

gradual increase in loss modulus until 70 °C, remaining constant until 65 °C, followed by a

sharp increase to 20 °C. The increase in G’ is an indication of gel formation (starch

gelatinisation) of the paste made from the commercial shrimp snack powder (predominantly

cassava flour).

111

The trends in G’ (storage modulus) values were slightly different over a temperature sweep of

20 °C to 90 °C for the horse mackerel extruded snack (Figure 5.5) and commercial shrimp

crackers (Figure 5.6). Although both the horse mackerel and commercial shrimp cracker had

low G’ values, during cooling, the commercial cracker showed an increase in both its stored

and elastic modulus values. G’ and G” increased in the commercial shrimp cracker

indicating that a weak gel developed which was absent in the horse mackerel extruded snack,

which showed no gelling. The low G’ and G” may be attributed to the denaturation of protein

caused by high temperatures during extrusion, particularly the destruction of myofibrillar

proteins that contribute to the gel network. This observation was confirmed by the low levels

of amino acids obtained after extrusion in this study.

Figure 5.6. A dynamic temperature sweep (20-90-20°C ) of a commercial shrimp

cracker

112

Large deformation tests using the texture analyser indicated that a higher force (g) (Table 5.6)

was required to compress or break the uncooked commercial shrimp cracker (Figure 5.7 b)

compared to the horse mackerel non-expanded extruded snack (Figure 5.7 a), but there was a

large degree of variability in the commercial samples. This result could be attributed the

dense nature of the commercial shrimp cracker. However, similar force (g) values were

obtained after frying of both uncooked samples (Figure 5.8 a and b) and there was no

significant difference in both products after frying. This suggests that despite the brittle

nature of the horse mackerel extruded non-expanded snack, an expanded fish snack with

similar hardness as fried commercial shrimp crackers can be obtained after frying when the

product expands further.

Table 5.6 Texture analyses of Uncooked Horse mackerel non-expanded extruded snack

made from Composite flour 2, Uncooked commercial shrimp cracker, Fried horse

mackerel extruded expanded snack made from Composite flour 2 and Fried

commercial shrimp cracker


2. Uncooked Horse mackerel non-expanded extruded snack made from Composite flour 2


3. Fried horse mackerel extruded expanded snack made from Composite flour 2 (20:20:30:30 horse

mackerel mince:maize:cassava:pearl millet)

Sample Force (g)

Uncooked Horse mackerel non-expanded extruded snack

Uncooked commercial shrimp cracker

Fried horse mackerel extruded expanded snack

679.5 ±143.9

2864.3 ±1038.1

947.9 ±209.9

Fried commercial shrimp cracker 990.5 ±333.6

113

4

a) b)

Figure 5.7. Texture profile of a) Uncooked Horse mackerel non-expanded extruded snack

made from Composite flour 2 (20:20:30:30 horse mackerel mince:maize:cassava:pearl millet)

b) Uncooked commercial shrimp cracker using the TAX-T2 Texture analyser

Figure 5.8 a) Fried Horse mackerel non-expanded extruded snack made from Composite

flour 2 (20:20:30:30 horse mackerel mince:maize:cassava:pearl millet) b) Fried commercial

shrimp cracker using the TAX-T2 Texture analyser

a) b)

114

Differential scanning calorimetry

The transition temperatures for the uncooked Horse mackerel non-expanded extruded snack

is illustrated in Table 5.7. The thermogram generated for uncooked Horse mackerel non-

expanded extruded snack indicated one transition that occurred at 21.7 Tm °C which

indicates that less energy was required as the starch was already gelatinised and proteins were

already denatured by extrusion. The enthalpy change (ΔH) value observed uncooked Horse

mackerel non-expanded extruded snack was 0.654 J/g and shows that less energy is required

after frying.

Table 5.7 Transition temperatures and enthalpy change (ΔH) for the uncooked horse

mackerel extruded snack (UHMES) made from Composite flour 2 and shrimp crackers

Sample Peak

Tm °C

Onset Temp

°C

Enthalpy

change J/g

Type of

reaction

Uncooked Horse

mackerel non-

expanded

extruded snack

Uncooked

commercial

shrimp cracker

21.7

30.9

16.6

26.4

0.654

1.59

Endothermic

Endothermic




115

The peaks observed for uncooked commercial shrimp cracker include peak 1 (Tm 26.4 –

30.9°C). The enthalpy change value observed in uncooked commercial shrimp cracker was

1.59 J/g is indicative of fewer conformational changes during processing of the dense and

compact shrimp cracker.

Table 5.8. Relative peak intensity of Raman spectra for regions 3200-700 cm-1 for

uncooked horse mackerel non-expanded extruded snack and commercial prawn

cracker


Peak assignment (wavenumber ± 2 cm-1) Uncooked

Extruded

Commercial

Shrimp Cracker

Trp (760) 0.6a ±0.1 0.9a ±0.3

Tyr (830, 0.7a ±0.1 0.7a ±0.1

855) 2.1a ±0.1 4.0b ±0.7

Helix C-C stretch,CH3, symmetric stretch

(937)

2.4a ±0.1 5.7b ±1.2

β-sheet type structure 990 0.4a ±0.1 0.5a ±0.1

Phe,ring band (1004) 1.6a ±0.1 1.2b ±0.2

Isopropyl antisymmetric stretch 3.9a ±0.2 5.1b ±0.4

CH stretch back bone (1128)

CH3 antisymmetric (aliphatic) CH3 rock 1.4a ±0.1 2.0b ±0.0

116





4. Data from an Independent Samples t- Test significance at 0.05

The phenylalanine band (1004 cm-1) was used as an internal reference for normalisation of

the spectra of the samples. The spectra for the commercial shrimp cracker (Table 5.8) showed

(aromatic) (1160)

β-sheet type (1239) 1.5a ±0.2 1.0b ±0.2

Amide III random coil (1245) 1.5a ±0.1 1.5a ±0.2

Amide III (1264) 2.1a ±0.1 1.9a ±0.4

Amide II (1320) 3.5a ±0.17 1.9b ±0.5

H band doublet from trp (1340) 5.2a ±0.2 5.2a ±0.4

(sh*,residue vibration) asp, glu, lys (1425) 2.1a ±0.2 0.4b ±0.3

Aliphatic groups, CH bend (1451) 3.4a ±0.4 2.4a ±0.6

Trp (1554) 0.3a ±0.2 0.1a ±0.0

Amide I (1660) 0.6a ±0.3 0.1a ±0.1

CH stretch, aliphatic (2940) 5.6a ±0.5 6.5a ±0.5

Shoulder (2888) 3.9a ±0.2 4.7a ±0.6

Shoulder (2976) (2969) 3.4a±0.3

3.9a ±0.3

3.8a ±0.3

4.5a ±0.3

117

a high intensity of the tyrosine doublet near 855 cm-1 and 830 cm-1 that relates to the number

of Tyr residues and the state of the hydrogen bonding of the phenolic OH group (Badii and

Howell, 2006). This could indicate hydrogen bonding between the molecules of the proteins

and starch in the commercial shrimp cracker. A decrease in the intensity ratio suggests the

role of the phenolic hydroxyl groups as strong hydrogen bond donors and state of the buried

tyrosine residues within the gel network .

The high vibrational peak at 937-940 cm-1 relates to a skeletal C-C stretching and is

proportional to the α-helix content. This band intensity decreased by heating which may be

the case when the commercial shrimp cracker is fried in oil. In general, a high intensity may

be attributed to the presence of a high protein concentration. However, an increase in peak

intensity may be due to starch encapsulating the protein during heat induced gelation.

The uncooked shrimp cracker had a higher relative intensity value of 5.11 in the 1128 cm-1

Raman band region compared to 3.86 in uncooked Horse mackerel non-expanded extruded

snack. This indicates a higher indole ring vibration of the tryptophan residues in the

uncooked shrimp cracker. Commercial shrimp cracker had a high relative intensity of 2.01

vibrational peak at 1160 cm-1 (Sarbon, Badii and Howell, 2015). Raman spectra in the

450−1900 cm-1 region relate to hydrophobic interactions by intensification of spectral bands

assigned to the CH and CH2 bending vibrations and a decrease in the intensity of bands

assigned to tryptophan residues in a nonpolar environment (Howell, Arteaga, Nakai, and Li-

Chan, 1999).

118

Amide II Raman bands are in the region of 1320 cm-1. The uncooked Horse mackerel non-

expanded extruded snack had a high relative intensity value of 3.51 compared to commercial

Shrimp Cracker of 1.92 in this region. C–H deformation (bending and stretching) mode of

aliphatic amino acids residues were seen in the 1400–1500 cm-1 (Sarbon, Badii and Howell,

2015). This band relates the interaction between hydrophobic and aliphatic residues. Polarity

and micro-environment affect the level of changes in C–H bending band intensity. These

results concur with the amino acid profile (Table 5.3) which showed higher levels of aromatic

and aliphatic amino acids in the Horse mackerel product compared with the commercial

shrimp cracker.

The uncooked Horse mackerel non-expanded extruded snack had a relative intensity value of

2.07 and commercial Shrimp Cracker 0.42 in the 1425 cm-1 region. The band around 1425

cm-1 relates to the COO symmetric stretch of the ionized carboxyl group vibration for

aspartic and glutamic acid residues (shoulder) and side chain vibrations of the imidazole ring

of histidine (Sarbon, Badii and Howell, 2015). This corresponds to the higher aspartic acid

and glutamic acid values of the amino acid profile of the uncooked Horse mackerel non-

expanded extruded snack compared with the shrimp cracker.

5.6 Conclusion

Extrusion enabled the combination of fish and cereals in varying proportions to improve the

nutritional properties of usual starchy snack products to provide an extruded fish snack with

a balanced nutritional content. Amino acid profile and differences in the protein bands

identified by Raman spectroscopy provided detailed information in the structure of proteins

119

that concurred and confirmed the higher protein quality of the horse mackerel extruded snack

product compared to the commercial shrimp cracker. Proteins were denatured and starch

gelatinised by the extrusion process as shown by low enthalpy values obtained by DSC. .

The trends in G’ (storage modulus) were slightly different over a temperature sweep of 20 °C

to 90 °C for the horse mackerel extruded snack and commercial shrimp crackers. Although

both the horse mackerel and commercial shrimp cracker had overall low G’ values, during

cooling, the commercial cracker showed an increase in both its stored and elastic modulus.

G’ and G” increased in the commercial shrimp cracker indicating further gelation or

gelatinisation on heating and cooling; this increase in G’ was absent in the horse mackerel

extruded snack, which was already denatured thus concurring with the DSC results. . The low

G’ and G” values may be attributed to the denaturation of proteins particularly the

myofibrillar protein gels caused by high temperatures during extrusion, and limited ability for

further gel network. This observation was confirmed by the low levels of amino acid obtained

in this study for the extruded product. However, after frying the rheological properties were

similar to the commercial shrimp cracker, indicating that it may be possible to enhance the

nutritional properties of the usual starchy snack products by the addition of fish that is high in

proteins and omega-3 fatty acids.

120

CHAPTER 6

121

CHAPTER 6

6. INFLUENCE OF RAW MATERIAL ON THE QUALITY OF THE EXTRUDED

HORSE MACKEREL SNACK AND A QUALITY COST MODEL

Aim

The aim of the investigation was to determine the influence of raw material on the quality of

an extruded fish-based snack product made from horse mackerel and local cereal mixtures

found in Namibia. Experimental results were applied to a mathematical model to determine

the relationship between the quality of raw material and the final extruded fish-based snack

product.

6.1 Introduction

Quality is an important consumer expectation which fishing industries in developing

countries should reach compliance with (Zugarramurdi, Parin, Gadaleta. Carizza and Lupin,

2004). Quality can be achieved through the implementation of a HACCP system which has

been shown to reduce quality related costs in seafood plants (Zugarramurdi, Parin, Gadaleta

and Lupin, 2007; Lupin, Parin and Zugarramurdi, 2010). The handling of the catch during

harvesting on board vessels affects the quality of raw material that is landed or which is to be

further processed. Gaping and bruising are examples of quality defects that arise during

filleting which affects the overall quality (Balaban, Şengör, Soriano and Ruiz, 2011).

These quality defects result in substantial economic losses for fishing companies as they are

unable to fetch premium prices for fish of reduced quality. The cost of these losses can be

122

quantified using a Quality Cost Model. This model factors in quality related costs such as

prevention costs, appraisal costs and failure costs (Feigenbaum, 1974). The Total Quality

Model comprises controllable and failure costs. Controllable costs include prevention (Pi)

and appraisal costs (Ai) which refer to those directly under the control of management for

decision making for product quality assurance. Failure costs comprise expenses arising from

defective products not conforming to quality requirements before and after reaching the

customer. Thus, Total Quality Cost is a summation of controllable and failure costs.

This research examined the effects of raw material quality on the quality of an extruded snack

food product derived from horse mackerel and was used to develop a Total Quality Cost

model of this nutritious snack food.


6.2.1 Materials

Whole horse mackerel fish (Trachurus capensis) was bought from local supermarkets in

Windhoek, Namibia. The fish were sealed in polyethylene bags and stored at -20°C until

analysed.

6.2.2 Methods

6.2.2.1 Fish sample preparation

Eight kilograms of frozen whole horse mackerel were divided into two sets. One set was used

for the determination of the influence of raw material quality on the sensory attributes of

whole fish and the second set was used for the determination of the influence of raw material

quality on yield. Each set was divided into three batches of frozen whole horse mackerel

123

fish, which were defrosted and stored in polyvinyl bins and kept in a cold room at 5°C for 0,

24 and 48 hours. Fish in the two batches, stored at 24 and 48 hours, were covered in ice,

whereas the Control (0 hours) contained no ice.

6.2.2.2 Yield

The whole fish as prepared in 6.2.2.1 were weighed and the weight was recorded. The fish

was gutted, deboned and filleted. The bones, offal and fillets were separated, and the weights

recorded were used to determine the percent yield. The fillets were used in the preparation of

the extruded snack as described in Chapter 5.

6.2.2.3 Sensory Evaluation of raw material with Quality Index Method

A six-member panel was screened and trained for visual colour and smell acuity. This panel

performed sensory evaluation on the fish raw material (described in section 6.2.2.1) using the

Quality Index Method (QIM). QIM is a demerit system used for fresh and frozen fish that

involves the evaluation and scoring of critical sensory parameters between 0 and 3.

Evaluation of the physical changes in the general appearance, and the appearance of skin,

eyes, gills, stiffness, belly, smell, odour and texture of raw fish. The Quality Index Method

scheme used in this study is presented in Table 6.1.

Table 6.1. Quality Index Method scheme (Adapted from Larsen et al., 1992 )

Quality Parameter Character Score (ice/seawater)

General Appearance Skin 0 Bright, shining

1 Bright

2 Dull

Bloodspot on gill cover 0 None

1 Small, 10-30%

2 Big, 30-50%

3 Very big, 50-100%

124

Stiffness 0 Stiff, in rigor mortis

1 Elastic

2 Firm

3 Soft

Belly 0 Firm

1 Soft

2 Belly burst

Smell 0 Fresh, seaweed/metallic

1 Neutral

2 Musty/sour

3 Stale meat/rancid

Eyes Clarity 0 Clear

1 Cloudy

Shape 0 Normal

1 Plain

2 Sunken

Gills Colour 0 Characteristic, red

1 Faded, discoloured

Smell 0 Fresh, seaweed/metallic

1 Neutral

2 Sweaty/slightly rancid

3 Sour stink/stale, rancid

Total of scores

6.2.2.4 Development of the horse mackerel fish-based snack

See Section 5.2.2.1.

6.2.2.5 Sensory evaluation of the final fish-based snack developed product

A six-member panel was screened and trained for visual colour and smell acuity. Sensory

evaluation using a 9-point scale was done on the extruded fish-based snack developed in

Section 5.2.2.1. The extruded fish-based snack developed was evaluated on appearance,

aroma, consistency, overall acceptability and a total sensory score was obtained.

125

6.2.2.6 Model and assumptions

The mathematical model as described by Zugarramurdi, Parin, Gadaleta and Lupin (2007)

was applied to the collected data (Tables 6.2 and 6.3) to develop the Total Quality Cost

Model (Figure 6.3). See Appendices 1.1, 1.2 and 1.3

Table 6.2. Quality, market and production parameters for extruded products

Quality

parameters:

YQm* = 0.706 kg extruded product/kg ingredients

XQm* = 10 kg product/h-worker

Nscp = 4

Nccp = 2

Market

parameters:

PQm = purchase price of raw material (US$/kg)

PQp = selling price for quality level Qp (US$/kg)

PQp* = 4 US$/kg Qp* = 0.8 US$/kg

Production

parameters:

d = 249 days per year

K = 18.5 kg/h

Ri = 75 kg/hour

S = 1 US$/h

IF = US$ 200 000 (2 t extruded fish/day)

L = 0.86/3.08 kg

126

Table 6.3. Relationships used to estimate quality costs components for extruded horse

mackerel

Type of cost Relationship Equation

Number

Controllable costs sub-

model

Prevention costs

P1. Design, development

and implementation of a

quality assurance plan

CP1 = 0.03*200000/(249*18.5)

(1)

P2. Quality training

programs for suppliers and

production personnel

CP2 = (0.05*(0.86/3.08))*Qp

(2)

P3. Hygiene and sanitation

of the plant

CP3 =

0.2*(0.86/3.08)*(1+(0.01*4))*Qp/0.842

(3)

P4. Preventive maintenance

and additional supervision

CP4 =

(0.002*200000)/(249*18.5+(0.08*(0.86/3

.08)*Qp

(4)

Appraisal Costs

127

A1. Receipt and control of

incoming material

CA1 = (1*1)/(C18*75*Qp)

(5)

A2. Sampling and

laboratory analysis

CA2 = 0.06*(0.86/3.08)*Qp

(6)

A3. In-process inspection CA3= 0.12*2*(0.86/3.08)*Qp (7)

Internal Failure Costs

F1. Scraps, reprocessing or

spoilage

CF1 = (4-PQp)*((1-Qp)2 (8)

F2. Low labour

productivity and low

process yield

CF2 = 1*((1/XQm)-

(1/10))+(PQm*((1/YQm)-(1/0.706)))

(9)

F3. Inefficient usage of

plant capacity

CF3 = (PQp*((0.706-YQm))/0.706) (10)

6.2.2.7 Statistical analyses

Total yield, QIM scores and results from the sensory evaluation of the final extruded product

were analysed on a dimensionless basis as described by Zugarramurdi et al. (2007).

Parameters were normalised to fit a scale of 0 to 1, with the value 0 representing the worst

quality and 1 representing optimum quality. Regression analysis was applied to all collected

data. Results were analysed using SPSS version 23 (IBM Corp. Released 2015. IBM SPSS

Statistics for Windows, Armonk, IBM Corp) and Microsoft Excel.

128


6.3.1 Influence of raw material quality

6.3.1.1 Extruded product quality

In this study, Cape horse mackerel (Trachurus capensis) which is a fatty fish was used

together with other raw materials e.g. maize, pearl millet and cassava snack to produce an

extruded horse mackerel snack. In order to analyse the influence of fish raw material quality

on extruded products it was assumed that only fish quality changed while the other

ingredients quality remained constant. The final product was a fried extruded horse mackerel

snack. The final product consisted of pieces of extrudate that were brown in colour, crispy

dry and slightly puffed texture with a total weight of 36 g in each aluminium pouch.

Figure 6.1. Influence of raw material quality on product quality in extruded Cape horse

mackerel

Since it was a fried product the primary packaging material used for the horse mackerel

extruded snack was aluminium foil pouches packed in carton boxes as secondary packaging.

y = 0.7515x + 0.0231R² = 0.6399

0

0.2

0.4

0.6

0.8

1

1.2

0 0.2 0.4 0.6 0.8 1 1.2 1.4

Pro

du

ct Q

ual

ity

(Dim

ensi

on

less

)

Raw Material Quality (Dimensionless)

129

Results of dimensionless analysis incorporating Sensory Evaluation data for the fried

extruded horse mackerel snack product are shown in Figure 6.1. The final correlation

between sensory evaluation data and product quality of the extruded horse mackerel snack is:

Qp = 0.7515 Qm + 0.0231 (R2 = 0.639, Student's t test, P < 0.01)

6.3.2 Filleting yield and productivity

The relationship between filleting yield and raw material quality was analysed. After

defrosting frozen whole horse mackerel, the fish was filleted, and the heads, guts and bones

Figure 6.2. Variation of filleting yield with raw material quality, dimensionless, for

horse mackerel

were discarded. In the fishing industry, the heads, guts and bones are typically used for

manufacturing fish meal. The weight of whole fish and fillet and the percentage weight loss

44

44.5

45

45.5

46

46.5

47

47.5

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

%Yi

eld

Raw Material Quality, Dimensionless

130

due to filleting was recorded. Results for yield as influenced by raw material quality are

shown in Figure 6.2.

6.3.3 Quality, market and production parameters

Actual figures were used to estimate quality costs. Selected values are shown in Table 6.4.

Table 6.4. Quality, market and production parameters for extruded products

Notes

Quality parameters:

YQm* = 0.86 kg extruded product/kg

ingredients

XQm* = 10 kg product/h-worker

Nscp = 4

Nccp = 2

Market parameters: PQm* = 0.8 US$/kg

PQp* = 4 US$/kg

Production parameters: d = 240 days per year

S = 1 US$/h 1

IF = US$ 200 000 (2 t extruded fish/day) 2

Fish proportion 30%

Notes for extruded products:

131

1. Payment rate is based on trained workers with GMP, HACCP and social charges

(Zugaramurdi et al, 2007)

2. Hypothetical plant

6.3.4 Quality costs estimation

Regression analysis was applied to the controllable costs, failure costs, and total quality costs

per unit of product for different levels of quality resulting from the application of the model.

Regression equations for the model were obtained for the extruded horse mackerel product,

polynomial curves were fitted to the model values by least squares estimations and were

plotted in Figure 6.3.

In food processing plants, the typical values associated with low quality products are 5% of

total preventions costs, 25% of appraisal costs and 70 failure costs in the context of the

overall total quality costs (Morgan, 1984). Failure to understand the relationships between the

components of cost, especially total quality cost and quality and failure to take preventive

actions, may be the reason why many plants incur major losses and lead to their downfall. In

contrast, should the aim be to produce a product of better quality, there is a need to increase

preventive efforts which in turn would help reduce defects improve overall quality (Sterling

1985; Sandholm, 1987).

6.3.5 Quality cost for extruded fish product

Data shown in Table 6.6 and equations in Table 6.3 were used to estimate the different

components of total quality costs. Results for extruded fish are shown in Figure 6.3. The

model was applied taking into consideration that a 30:70 proportion of fish:cereal was used

132

as described in Chapter 5. The analysis was performed keeping constant the quality of cereals

and other ingredients while varying the fish quality.

Figure 6.3. Total quality costs, failure costs and controllable costs as a function of

product quality for the extruded horse mackerel fish snack

Regarding the quality cost estimation for a hypothetical plant that manufactures extruded

horse mackerel snacks, results indicated that product quality level increases from fairly poor

to very good depending on the raw material quality used during processing. Controllable

costs increase from less than 1% to 80% of Total Quality Cost as the quality of raw material

deteriorates. Failure costs decrease from 80% to 44% of Total Quality Cost when good

quality material is used in developing the extruded horse mackerel snack. The minimum TQC

is reached at 80% of the maximum product quality (Zugarramurdi, 2013). The results indicate

that horse mackerel can be stored up to a maximum of 24 hours at 5°C for it to be used in

0

0.2

0.4

0.6

0.8

1

1.2

0 0 . 2 0 . 4 0 . 6 0 . 8 1 1 . 2

PR

OD

UC

TIO

N C

OST

US$

/KG

PRODUCT QUALITY DIMENSIONLESS

Total Controllable CostTotal Failure CostTotal Quality Cost

133

processing an extruded horse mackerel snack. Beyond this period will result in poor horse

mackerel raw material quality level that directly affects the overall quality of the horse

mackerel extruded snack. In addition, labour costs increase and production capacity

decreases.

6.4 Conclusions

The quality costs estimation was based on experimentation carried out using a laboratory

scale single screw extruder and using assumptions based on the literature and the results

obtained in the study should only be treated as preliminary results. A realistic approach

would be to validate and verify the findings of this study in an actual plant that develops

snacks. This manufacturing plant should have a quality management system in place or at

least is implementing Good Manufacturing Practices and Sanitation Standard Operating

Practices. Results showed that if poor quality fish is used in the processing of the extruded

horse mackerel snack, failure costs could be over 95 % of total cost. Correlations between

product quality and raw material quality for the horse mackerel extruded snack can result in

the reduction of failure costs to below 20% of total quality costs. Investing in quality

management systems that prolong the optimum quality of raw material can keep controllable

costs low to achieve more than 80% of Total Quality Cost.

134

CHAPTER 7

135

CHAPTER 7

7. GENERAL DISCUSSION

Fisheries is the second largest industry in Namibia (Lange, 2003). In Namibia, the current

practice after catching Cape horse mackerel (Trachurus capensis) is to freeze and package

fish whole in cartons onboard fishing vessels. These cartons are then translocated from

fishing vessels to cargo ships which transport the frozen whole cape horse mackerel fish to

the markets. In the past, the juvenile Cape horse mackerel fish that were small and did not

meet the export standards were discarded by fishing vessels into the sea. This had become an

environmental problem and authorities compelled fishing companies involved in harvesting

horse mackerel to bring the fish ashore. Landing fish of such small size and low in value is a

loss to fishing companies. To recover some of the cost, the juvenile horse mackerel which has

a high protein content is used to manufacture fish meal (Mukumangeni, and Wiium, 2006).

The main objective of the study was to investigate the physicochemical and biochemical

properties of juvenile Cape horse mackerel (Trachurus capensis) and to develop a value-

added product like an extruded horse mackerel snack. Extrusion technology can be used to

add value to Namibia’s underutilised fish resources like juvenile Cape Horse mackerel by

manufacturing edible fish products for human consumption instead of utilising it solely for

fish meal. In this study, a low-cost single screw extruder designed by the Dienst

Landbouwkundig Onderzoek (DLO) in the Netherlands and manufactured by the Safe World

FoodTech Company in Malaysia was used to develop a horse mackerel extruded snack from

juvenile Cape horse mackerel in order to make this rich source of protein and fish oil

available to consumers.

136

7.1 Proximate and biochemical composition of Cape horse mackerel (Trachurus

capensis), pearl millet (Pennisetum glaucum L. Br, Maize (Zea mays) and Cassava

(Manihot esculenta)

In Chapter 2, the nutritional and physico-chemical properties of the raw material (horse

mackerel fish mince, pearl millet, maize and cassava flours) were characterised. The protein

content of horse mackerel ranged between 16.4% (whole fish) and 20% (edible portion).

These values were lower than the 21% reported by Badii and Howell (2006) in Atlantic

Horse mackerel. The difference in protein contents can be due to variations in the location

where fish was harvested. This study reports protein content of both whole fish and the edible

portion flesh. The former would be relevant for canned horse mackerel, whereas the latter

explains the 18.3% protein content found in the extruded horse mackerel snack (Chapter 5).

Extrusion with cereals lowered the protein content of the final extruded snack. This is

expected since cereals have a lower protein content. Although protein is denatured by the

high temperatures during extrusion most amino acids are still available.

The high fat content of 4.3% in Cape horse mackerel is typical for pelagic fish. Seasonality,

feed intake, spawning and water temperature affect the lipid percentage in medium fat pelagic

fish such as horse mackerel, which subsequently this affects the fatty acid profile of the fish.

The fatty acid composition of horse mackerel indicated high levels of oleic acid, C18:1

(31.3%), palmitic acid, C16:0 (28%) and stearic acid, C18:0 (12.4%). This is in line with the

observations of Metillo and Aspiras-Eya, (2014) who reported that high levels of palmitic and

stearic acids are found in warm water fish compared with cold water fish. Cape horse

mackerel like Atlantic horse mackerel have properties similar to warm water fish (Badii and

Howell, 2006).

137

Polyunsaturated fatty acids such as eicosapentaenoic acid (EPA) and docosahexaenoic acid

(DHA) are found in high levels in fish and are reported to lower the risk of cardiovascular

diseases and are important in the development and functioning of the central nervous system.

In this study, docosahexaenoic acid (DHA) was not detected in Cape horse mackerel.

Docosahexaenoic acid (DHA) is sensitive to oxidation (Lyberg, Fasoli and Adlercreutz,

2005). It is hypothesized that after cold lipid extraction the fish oil sample needs to be

injected immediately into the gas chromatograph. Saeed and Howell (1999) reported 14.84%

DHA in extracted lipids in Atlantic horse mackerel. Thus, the absence of DHA in the Cape

horse mackerel may have been be due it undergoing oxidation before it could be analysed by

gas chromatography.

Lipids, however, may store mercury and polychlorinated biphenyls (PCBs) which are toxic.

In this study the presence of toxic substances was not investigated. In a previous EU project

several species of fish including Cape horse mackerel from Namibia were tested for PCBs

(Sarkadei, 2005). In their study, very low levels of PCBS were found in the fish from

Southern hemisphere countries, which were lower than those found in the Northern

hemisphere species, and all were well below the allowable limits. However, as stated earlier

it is recommended that vulnerable populations like pregnant women and children limit

consumption of fatty fish to twice a week.

Characterisation of the water and salt soluble proteins by polyacrylamide gel electrophoresis

identified protein bands with molecular weights of 17,000 and 95,000 daltons reflecting the

presence of salt soluble myofibrillar proteins myosin and actin as well as smaller

138

sarcoplasmic proteins. Cape horse mackerel had showed the full complement of amino acids

but was rich in glutamic acid (9.3%), lysine (8.9%) and glycine (7.7%) levels. Glutamic acid

is required in the neurotransmitters of the central nervous system of the human body. Lysine

is an essential amino acid and an absence of essential amino acids in the human diet may lead

to symptoms of Protein Energy Malnutrition. Glycine is a precursor for protein arrangement.

The amino acid profile of fish typically has higher levels of sulphur containing essential

amino acids than plant proteins (Usydus,Szlinder-Richert and Adamczyk, 2009), hence the

ideal is to have a product that includes a balance between plant and animal proteins. It is

recommended that that cereal based diets should be supplemented with high quality protein

food such as fish, meat, egg or dairy products. Therefore in this study, from a nutritional

point of view, fish amino acids, fatty acids and cereals (starch) complemented each other

when combined and cooked into a wholesome product through extrusion. Extrusion cooking

takes place under high pressure and temperature and can produce a variety of food products

in a unit process combining cooking and forming at low cost. Extrusion enabled the

combination of fish and cereals in varying proportions to improve the nutritional properties of

usual starchy snack products to provide an extruded fish snack with a balanced nutritional

content.

7.2 Rheological, DSC, FT-Raman spectroscopy of horse mackerel fish mince, pearl

millet, maize, cassava flours, product development and mixtures

In chapter 3, the rheological and functional properties of raw material (horse mackerel fish

mince, pearl millet, maize and cassava flours) used to make the extruded horse mackerel fish

snack were first individually determined. Rheology revealed the effects of temperature on the

proteins in fish and cereals. In fish, storage modulus (G’) values increased in the temperature

range of 20°C to 46°C and at a later temperature just before 90°C, whereas in cereals the

139

storage modulus (G’) values decreased. These temperatures correlates to the temperature

range when the denaturation of myosin and sarcoplasmic proteins occurs as confirmed by

results of differential scanning colorimetry. Differential Scanning Calorimetry supported the

occurrence of protein unfolding and denaturation as shown the transition temperatures and

correlated with the G’/G” values shown by rheology. Low enthalpy values obtained by DSC

indicated protein denaturation. The third increase in G’ values at approximately 60°C is

attributed to the denaturation of actin which also correlated with the third transition peak

observed in the differential scanning colorimetry results.

The decrease in G’ values trend observed in small deformation rheology of cereals is

attributed to the high amounts of starch (Petrofsky and Hoseney, 1995). Since most flours

were mostly already in a solid paste state, storage modulus values G’ were always higher than

loss modulus (G”). An increase in storage modulus G’ values was attributed to the melting of

crystalline regions of starch granules and the swelling of starch during gelatinisation resulting

in the formation of a starch granular network (Alcázar-alay and Almeida Meireles,2015).

Consequently, starch gelatinisation by the extrusion process resulted in low enthalpy values

obtained by differential scanning colorimetry.

7.3 Product Development and Mixtures

The overview of the individual rheological and functional properties of the raw material in

Chapter 3, formed the basis used to identify and formulate the three mixtures of cereals and

fish in Chapter 4, as models for the extruded product. The approach taken was to determine

the effect of varying amounts of fish in the mixture of the three flours and to select one flour

that would compensate the difference. Pearl millet was selected as the cereal that would vary

in proportion in the mixture with varying fish levels because it was shown in Chapter 2 that it

140

had a higher protein and amino acid score among the three flours. Furthermore, it is the staple

food of Namibia and thus it is more readily available than cassava flour that is imported. The

other two flours, maize and cassava, were kept at constant proportions. Cassava and maize

are widely used in the development of snack foods and thus their functional properties are

well understood and widely reported (Assumpção Fiorda et al., 2015; Serna-Saldivar, S.O.

and Rooney, 2015). Pearl millet, however, is not typically used in the development of snacks.

Mixtures of fish and cereals with varying composition can be used to formulate extruded

snack products of similar textural properties as shown in Chapter 5.

As stated, gelatinised starch in cereals, affected the formation and stability of the gel network

that manifested in the increase in G’ and G” values. Composite flour 2 consisting of a

20:20:30:30 horse mackerel, maize flour, pearl millet flour and cassava flour respectively

showed the best starch and protein interactions as evidenced by thermal analyses and

rheological properties. This is the rationale for choosing the above mixture in the

development of the horse mackerel extruded snack in Chapter 5.

The trends in G’ (storage modulus) values were slightly different over a temperature sweep of

20°C to 90°C for the horse mackerel extruded snack and commercial shrimp crackers.

Although both the horse mackerel and commercial shrimp cracker had overall low G’ values,

during cooling, the commercial cracker showed an increase in both its stored and elastic

modulus. G’ and G” values increased in the commercial shrimp cracker indicating further

gelation or gelatinisation on heating and cooling. A similar increase in G’ values was absent

in the horse mackerel extruded snack, which was already denatured on heating; this result

concurred with the DSC results. The low G’ and G” values may be attributed to the

141

denaturation of proteins particularly the myofibrillar proteins, caused by high temperatures

during extrusion, and limited ability for further gel network. This observation was confirmed

by the low levels of amino acids obtained in this study for the extruded product compared

with the uncooked fish.

However, the amino acid profile and differences in the protein bands identified by Raman

spectroscopy provided detailed information on the structure of proteins that occurred and

confirmed the higher protein quality of the horse mackerel extruded snack product compared

to the commercial shrimp cracker.

After frying the extruded fish-cereal product, the rheological properties in terms of texture

were similar to the commercial shrimp cracker, indicating that it may be possible to enhance

the nutritional properties of the usual starchy snack products by the addition of fish that is

high in proteins and omega-3 fatty acids.

7.4 Extrusion

In this study, there were limitations observed in the design of the extruder. First, the flights of

the screw in the metering section were not sufficiently segmented for it to build high levels of

pressure required to produce highly expanded cereal and snack foods. Consequently, dense

and poorly expanded extruded products were initially formed. The design was more of a low

shear cooking extruder used for meat like snacks instead of a high-pressure forming extruder

typically used for cereal and snack foods where high pressures are required to achieve

expansion.

142

In this study, the extrusion method as described by Cheow Chong Seng (2013) was modified.

The modification related to changing the extrusion processing parameters meant for a ready

to eat first generation expanded snack to parameters that would result in a semi cooked non-

expanded extrudate of molten mass, that exits the barrel of the extruder in a semi-solid state

that can be rolled. After rolling, the flattened extrudate was dried for twenty-four hours at

50oC. The horse mackerel fish snack developed in this study is therefore classified as a third

generation snack. Third generation snacks are moisture stable products that are semi-cooked,

and that are subsequently expanded by frying in hot oil, microwaving or puffing.

Another disadvantage observed in the design of the extruder was that it had no provisions of a

mechanism to add water in the various sections of the extruder barrel. Altering the moisture

content of the extruded feed passing through the extruder barrel allows for changes in the

textural properties of extrudates. The current design of the extruder can only utilise the

moisture content in the feed. Furthermore, the design of the single screw extruder with a

closed barrel system requires the disassembling of the breaker plate and the removal of the

extruder screw after each run when cleaning, which was an arduous task.

Lastly, the design of the cooling system of the extruder utilised a low density polyethylene

tubing to carry water across the cooling system of the extruder. Heat generated from the

heating bands caused the tubing to rupture. A more robust alternative would have been to use

stainless steel or copper tubing to transport the water across the cooling system. An

ineffective cooling system created difficulties in regulating and maintaining set temperatures

and created heat fluctuations especially with additional heat generated by friction during the

extrusion of raw material.

143

7.5 Influence of raw material on the quality of fish products

In Chapter 6, The Quality Index Method was found to be a good tool to evaluate the raw

material quality of horse mackerel. QIM provides a holistic quality assessment of fish. The

findings of this section are critical for the fishing industry as well as for any entity that

ventures into commercialising the extruded horse mackerel snack. Horse mackerel should be

processed immediately into the extruded snack once it has reached a defrosted state. Using

raw material that is stored at chilling temperatures of 5 °C and ice for 24 hours slightly

delayed the deterioration of horse mackerel fish. Such raw material would produce a product

that would be considered sensorially barely acceptable. Horse mackerel stored for an

additional 24 hours in ice in a chiller would result in an extruded snack that would be

regarded as completely unacceptable.

The reason for horse mackerel fish rapid deterioration is due to its high protein, fat, amino

acids and fatty acid contents as reported in Chapter 2. A high protein and amino acid content

favours the growth of spoilage micro-organisms. Micro-organisms produce enzymes that

cause the deamination of proteins. The high PUFA fat content makes the fish prone to

oxidative rancidity. Despite the high temperatures involved in extrusion, fish of compromised

quality if passed on to the final product would result in an extruded product of inferior

quality.

Furthermore, the total filleting yield of horse mackerel is reduced if it is not processed

immediately after defrosting. Muscle deterioration and shrinkage occurs from prolonged

storage in an unfrozen state, consequently, making the separation of meat from bones during

filleting problematic; this can result in loss of fish and increased waste material. The practical

144

implication therefore relates to how long the horse mackerel can be kept in storage bins of the

chiller prior to processing, this is the information plant managers typically rely on.

The Zugarramurdi and Lupin Total Quality Cost Model was a good tool to use to evaluate

and cost such losses. Their model was applied to actual existing seafood processing plants

where actual data is readily available. In this study, processing data was from a laboratory

scale extruder and several assumptions were made for a hypothetical plant that would process

the extruded snack. Some assumptions like labour productivity were based on the reported

literature. In addition, only two raw material parameters were used to determine the

regression model. Ideally, more physico-chemical data such as Total Volatile Bases Nitrogen

(TVB-N), Thiobarbituric acid (TBARS) and microbiology would strengthen the regression

analyses. It is for these reasons that the Total Quality Cost Graph in Chapter 6, slightly

differed with those reported in previous studies where the Zugarramurdi and Lupin Total

Quality Cost Model was applied. However, the results clearly indicated the influence of raw

material quality on the quality of final extruded product and total cost. The findings were in

line with studies on fatty fish and other dried and extruded products.

7.6 CONCLUSIONS

• The nutritional and physico-biochemical properties of horse mackerel and local cereals

such as pearl millet, maize and cassava flours found in Namibia and used in the

development of an extruded horse mackerel snack were assessed. The results of this study

confirmed that Cape horse mackerel is rich in monounsaturated and polyunsaturated fats

and has a good amino acid profile. Horse mackerel in its raw form has a high nutritional

significance which is more evident when it is added to cereal flours.

145

• The functional and rheological properties of horse mackerel and local cereals such as

pearl millet, maize and cassava flours found in Namibia were characterised and used in

the identification and formulation of fish and cereal mixtures that were subjected to

extrusion in order to develop the extruded horse mackerel snack. Varying the composition

of the mixtures based on their functional and rheological properties resulted in extruded

products with similar textural properties to the model systems.

• A composite flour consisting of 20% horse mackerel, 20% maize flour 30% pearl millet

flour and 30% cassava flour showed the best starch and protein interactions as evidenced

by thermal analyses and rheological properties; this formulation was used in the

development of the horse mackerel extruded snack.

• The extrusion process was optimised and an extruded snack made of horse mackerel and

cereals with similar functional properties as commercial shrimp crackers was developed

successfully. The horse mackerel extruded snack and commercial shrimp crackers

differed slightly as shown in the trends in G’ (storage modulus) over a temperature sweep

of 20°C to 90°C. This increase in G’ and G” values characterised the formation and

stability of the gel network during starch gelatinisation.

• Low G’ values were observed in both the horse mackerel extruded snack and commercial

shrimp cracker. However, in the commercial crackers an increase in the stored and elastic

modulus was found during cooling. In the commercial shrimp cracker, the G’ and G”

increased indicating further gelation or gelatinisation on heating and cooling. In the horse

146

mackerel extruded snack this increase in G’ was absent, indicating that the fish product

had undergone denaturation; this concurred with the DSC results. Differential Scanning

Calorimetry Tm and enthalpy change results supported the occurrence of protein

unfolding and denaturation and the transition temperatures correlated well with the G’/G”

values.

• Furthermore, low G’ and G” values were due to the denaturation of proteins particularly

the myofibrillar proteins caused by the high temperatures during extrusion thus limiting

further gel network formation. Low enthalpy values obtained by DSC indicated starch

gelatinisation and protein denaturation by the extrusion process.

However, after frying, the rheological properties were similar to those of the commercial

shrimp cracker, indicating that it may be possible to enhance the nutritional properties of

the usual starchy snack products by the addition of fish that is high in proteins, essential

amino acids and omega-3 fatty acids.

• The higher protein quality of the horse mackerel extruded snack product compared to the

commercial shrimp cracker was shown by the horse mackerel extruded snack amino acid

profile and detailed information in the structure of proteins and differences in the protein

bands as revealed by Raman spectroscopy.

• The raw material quality used in the development of an extruded horse mackerel snack

influenced its final quality. The losses in quality were quantified with a Total Quality

Cost model. Implementing Good Manufacturing Practices and Sanitation Standard

Operating Practices in food plants can extended the quality of raw material which in turn

147

extends the final product quality. If poor quality fish is used in the processing of the

extruded horse mackerel snack, failure costs could be as high as 95 % of the total quality

cost. The relationship between product quality and raw material quality for the horse

mackerel extruded reduces failure costs to below 20% of total quality costs, depending on

the quality of raw material used. Investing in preventative actions such as quality

management systems can prolong the optimum quality of raw material to keep

controllable costs low and to achieve more than 80% of Total Quality Cost.

7.7 Further work

• It is suggested that scaling up studies be undertaken to investigate the requirements of

producing the extruded horse mackerel snack on a commercial scale. This would enable

small medium enterprises to create a business venture from utilising juvenile Cape horse

mackerel.

• The nutritional study results revealed that the extruded snack had a high fat content. It is

recommended that further studies should be aimed at identifying natural antioxidants

from seaweeds that could be used as a preservative of the horse mackerel snack.

Seaweeds are abundant on the Namibian coastline and are a rich source of polyphenolic

compounds that can act as antioxidants.

• Shelf life studies could be undertaken to investigate the stability of the extruded horse

mackerel snack with a focus on maintaining its quality over a specific period before

deterioration occurs. Such a study could be aimed at determining the preservation

capabilities of various packaging materials.

148

• In this study, numerous challenges were experienced with using a single screw extruder.

It is suggested that a study be undertaken to investigate the quality of products processed

using a twin-screw extruder.

• This study investigated the use of horse mackerel which is a marine fish species. Another

suggestion would be to investigate the use of fresh water fish species like tilapia in

developing an extruded snack. These species are found naturally in rivers and other water

bodies or are cultured in the same areas where the cereals such as pearl millet and maize,

used in developing the extruded horse mackerel, are grown. Should cultured fish be used

in the development of an extruded fish snack, there would be a need to screen these fish

for toxic Polychlorinated biphenyls (PCBs), dioxins and other contaminants to ensure that

the cultured fish are safe for human consumption.

149

REFERENCES

150

8. REFERENCES

Akawa, T. and Nashima, F.P. 2014. The role of fisheries management in promoting fishery

development: An Analysis of the Namibian Horse Mackerels (Trachurus capensis) Journal

of Agriculture and Biodiversity Research 9-13.

Alcázar-alay, S.C. and Almeida Meireles, M.A. 2015. Physicochemical properties,

modifications and applications of starches from different botanical sources. Food Sci.

Technol 35(2), 215-236.

Ali, M.A.M., El Tinay, A.H. and Abdalla, A.H.2003. Effect of fermentation on the in vitro

protein digestibility of pearl millet. Food Chemistry 80 (1) 51–54.

Alloncle, M. and Doublier, J-L. 1991. Viscoelastic properties of maize starch/hydrocolloid

pastes and gels. Food Hydrocolloids 5(5), 455-467.

Annor, G.A., Marcone, M., Bertoft, E., Seetharaman, K.2014. Physical and molecular

characterization of millet starches. Cereal Chemistry 91(3), 286-292.

Annor, G.A., Tyl, C., Marcone, M., Ragaee, S. and Marti, A. 2017. Why do millets have

slower starch and protein digestibility than other cereals? Trends in Food Science and

Technology 66, 73-83.

Anton, A.A., Fulcher, R.G., Arntfield, S.D. 2009. Physical and nutritional impact of

fortification of corn starch-based extruded snacks with common bean (Phaseolus vulgaris L.)

flour: Effects of bean addition and extrusion cooking. Food Chemistry 113, 989–996.

AOAC. (2016). Official methods of analysis of AOAC international (20th ed.). Virginia,

USA: Association of Official and Analytical Chemists International.

151

Arun Kumar, T.V., Samuel, D.V.K., Jha, S.K. and Sinha, J. P. 2015. Twin Screw Extrusion

of Sorghum and Soya Blends:A Response Surface Analysis. J. Agr. Sci. Tech. 17, 649-662.

Assumpção Fiorda,F. Soares Jr.,M., Alves da Silva, F., Melo Araújode Moura, C., and Eiras

Grossmann, M.V. 2015. Physical quality of snacks and technological properties of pre-

gelatinized flours formulated with cassava starch and dehydrated cassava bagasse as a

function of extrusion variables. LWT - Food Science and Technology 62(2), 1112-1119.

Aubourg, S.P., Rodriguez, A. and Gallardo, J.M. 2005. Rancidity development during frozen

storage of mackerel (Scomber scombrus): effect of catching season and commercial

presentation. Eur. J. Lipid Sci. Technol. 107, 316–323.

Badea,E., Gatta, G.D. and Budrugeac,P. 2011.Characterisation and evaluation of the

environmental impact on historical parchments by differential scanning calorimetry. Journal

of Thermal Analysis and Calorimetry 104(2), 495-506.

Badii, F. and Howell, N.K. 2001.A comparison of biochemical changes in cod (Gadus

morhua) and haddock (Melanogrammus aeglefinus) during frozen storage. Journal of Science

in Food Agriculture 82, 87-97.

Badii, F. and Howell, N.K. 2002. Effect of Antioxidants, Citrate, and Cryoprotectants on

Protein Denaturation and Texture of Frozen Cod (Gadus morhua). J. Agric. Food Chem. 50,

2053-2061.

Badii, F., and Howell, N. K. 2003. Elucidation of the effect of formaldehyde and lipids on

frozen stored cod collagen by FT-Raman spectroscopy and differential scanning calorimetry.

J. Agric. Food Chem. 51, 1440-1446.

Badii, F., and Howell, N.K. 2006. Fish gelatin: Structure, gelling properties and interaction

with egg albumin proteins. Food Hydrocolloids 20, 630-640.

152

Baeza, R.I., Carp, D.J., Pérez,O.E. and Pilosof, A.M.R. 2002. κ -Carrageenan—Protein

Interactions: Effect of Proteins on Polysaccharide Gelling and Textural Properties. LWT -

Food Science and Technology 35(8), 741-747

Balaban, M.O., Sengör, G.F.U., Soriano, M.G., Ruiz, E.G., 2011. Quantification of gaping,

bruising, and blood spots in salmon fillets using image analysis. J. Food Sci. 76, 291–297.

Bandarra, N.M., Batista, I. Nunes, M.L. and Empis, J.M. 2001.Seasonal variation in the

chemical composition of horse-mackerel (Trachurus trachurus) Eur Food Res Technol 212,

535–539.

Barroca, M.J. 2016. Breakfast cereals. In: Engineering Aspects of Cereal and Cereal-Based

Products (1st Edition).R.de Pinho Ferreira Guine and P.M.dos Reis Correia (Eds), CRC

Press, pp 201-208.

Bauleth-D'Almeida , G., Krakstad, J-O. and Kanandjembo. A. 2001. Comparison of horse

mackerel length frequencies obtained from research vessels and commercial midwater

trawlers: implications for biomass estimation. S. Afr. J. mar. Sci. 23, 265–274

Bhupender, S.K., Rajneesh, B., Baljeet S.Y. 2013.Physicochemical, functional, thermal and

pasting properties of starches isolated from pearl millet cultivars International Food

Research Journal 20(4), 1555-1561.

Bidlingmeyer, B.A., Cohen, S.A. and Tarvin, T.L. 1984.Rapid analysis of amino acids using

pre-column derivatization. Journal of Chromatography 336, 93-104.

Boyer, D.C. and Hampton, I. 2001. An overview of the living marine resources of Namibia.

South African Journal of Marine Science 5-30.

Bradford, M. 1976. Protein reaction with dyes. Anal. Biochem. 72, 248–251.

153

Bruntrup,M. and Herrmann,R. 2012. In: Global Value Chains: Linking Local Producers from

Developing Countries to International Markets, M.P.Van Dijk, and J.Trienekens (Eds)

Amsterdam University Press p 97.

Ceballosa, H.,Sáncheza,T.,Cháveza, A.L., Iglesias, C., Debouck, D.,Mafla, G.,and Tohme, J.

2006. Variation in crude protein content in cassava (Manihot esculenta Crantz) roots. Journal

of Food Composition and Analysis 19, 589–593.

Celik, M. 2008. Seasonal changes in the proximate chemical compositions and fatty acids of

chub mackerel (Scomber japonicus) and horse mackerel (Trachurus trachurus) from the

north eastern Mediterranean Sea. International Journal of Food Science and Technology 43,

933–938.

Chaiyakul, S., Jangchud, K., Jangchud,A., Wuttijumnong, P. and Winger, R. 2009. Effect of

extrusion conditions on physical and chemical properties of high protein glutinous rice-based

snack LWT - Food Science and Technology 42, 781–787.

Chan, J. K. Gill, T. A. and Paulson, A. T. 1992. The dynamics of thermal denaturation of fish

myosins. Food Research International 25, 117-123.

Chandna,M. and Matta, N. K. 1990. Characterization of pearl millet protein

fractions.Phytochemistry 29(11), 3395-3399.

Chang, C.L. and Decklebaum,R.J. 2013. Omega-3 fatty acids: mechanisms underlying

'protective effects' in atherosclerosis. Curr Opin Lipidol 24(4), 345–350.

Charles,A.L.,Sriroth,K., and Huang,T. 2005. Proximate composition, mineral contents,

hydrogen cyanide and phytic acid of 5 cassava genotypes. Food Chemistry 92, 615–620.

Charoenkul, N., Uttapap, D., Pathipanawat, W. and Takeda, Y. 2011. Physicochemical

characteristics of starches and flours from cassava varieties having different cooked root

textures. LWT - Food Science and Technology 44, 1774-1781.

154

Chaudhary, A.L., Miler, M., Torley, P.J., Sopade, P.A., Halley, P.J. 2008. Amylose content

and chemical modification effects on the extrusion of thermoplastic starch from maize.

Carbohydrate Polymers 74, 907–913.

Cheng, H. and Friis, A. 2010. Modelling extrudate expansion in a twin-screw food extrusion

cooking process through dimensional analysis methodology. Food and Bioproducts

processing 88, 188–194.

Cheow Chong Seng. 2013. 18 Month report Securefish Project, presented June 2013.

Cimmaruta, R., Bondanelli, P., Ruggi, A. and Nascetti, G.2008. Genetic structure and

temporal stability in the horse mackerel (Trachurus trachurus). Fisheries Research 89, 114–

121.

Coetzee, J. C., Misund O. A and Oechslin, G.2001. Variable spatial structure of schooling

pelagic fish off Namibia: implications for acoustic surveys. South African Journal of Marine

Science 23(1), 99-109.

Creighton, T.E (1993) Proteins: Structures and molecular properties. Publ. W H Freeman and

Company, New York, 2nd Edition.

Demeke, K.H. 2018. Nutritional Quality Evaluation of Seven Maize Varieties Grown in

Ethiopia Biochemistry and Molecular Biology 3(2), 45-48

Deng,J., Li,K., Harkin-Jones, E., Price, M., Karnachi, N. ,Kelly, A.,Vera,Sorroche,

J.,Coates,P.,Brown,E. and Fei,M. 2014. Energy monitoring and quality control of a single

screw extruder. Applied Energy 113, 1775-1785.

De Souza, P.C. and Bonilla-Rodriguez., G.O. 2007. Fish hemoglobins. Brazilian Journal of

Medical and Biological Research 769-778.

155

Ejeta, G., Hassen, M. M. and Mertz, E. T.1987. In vitro digestibility and amino acid

composition of pearl millet (Pennisetum typhoides) and other cereals. Proc. Natl. Acad. Sci.

84, 6016-6019

Erkan, N., Selçuk, A. and Özden, Ö. 2010. Amino Acid and Vitamin Composition of Raw

and Cooked Horse Mackerel. Food Anal. Methods 3, 269–275

Eskin, N.A.M., Aliani, M and Shahidi, F. 2013. Biochemistry of Foods 127-185.

Emerson, I.A. and Amala,A. 2017. Protein contact maps: A binary depiction of protein 3D

structures. Physica A 465, 782–791.

Eymard, S., Baron, C.P. and Jacobsen, C. 2009. Oxidation of lipid and protein in horse

mackerel (Trachurus trachurus) mince and washed minces during processing and storage.

Food Chemistry 114, 57–65.

Farvin, K.H.S., Grejsen, H.D. and Jacobsen C. 2012. Potato peel extract as a natural

antioxidant in chilled storage of minced horse mackerel (Trachurus trachurus): Effect on

lipid and protein oxidation. Food Chemistry 131, 843–851.

Fauconneau, B., Grayt, C. and Houlihant, D. F. 1995. Assessment of individual protein

turnover in three muscle types of rainbow trout. Comp. Biochem. Physio 45-51.

Feigenbaum, A.V. (1974). Total quality control (1st ed.). McGraw-Hill, New York.

FAO.1989. Food and Agriculture Organization of the United Nations. Yield and nutritional

value of the commercially more important fish species. FAO Fisheries Technical Paper 309.

Food and Agriculture Organisation. 1995. Sorghum and millets in human nutrition. p 47.

Food and Agriculture Organisation. 2012. The state of food insecurity in the world. p 47.

156

Food and Agriculture Organisation.2013. Species Fact Sheets Trachurus capensis (Castelnau,

1861). Available at http://www.fao.org/fishery/species/3105/en (Accessed on: 18 August

2015).

Fraser, O.P. and Sumar, S.1998. Compositional changes and spoilage in fish (part II) -

microbiological induced deterioration. Nutrition and Food Science 98(6), 325-329.

Frydenberg, R.P., Hammershøj, M., Andersen, U., Greve, M.T. and Wiking, L.2016. Protein

denaturation of whey protein isolates (WPIs) induced by high intensity ultrasound during heat

gelation. Food Chemistry 192, 415–423.

Gaonkar, A.G. and McPherson, A. 2005. Ingredient Interactions: Effects on Food Quality,

Second Edition CRC Press.

García-Arias, M.T., Álvarez Pontes, E., Garcíá-Linares, M.C., García-Fernández, M.C. and

Sánchez-Muniz, F.J.2003. Cooking–freezing–reheating (CFR) of sardine (Sardina

pilchardus) fillets. Effect of different cooking and reheating procedures on the proximate and

fatty acid compositions. Food Chemistry 83, 349–356

Gill, T. A., Chan, J. K., Phonchareon, K. F. and Paulson, A. T. 1992. Effect of salt

concentration and temperature on heat-induced aggregation and gelation of fish myosin. Food

Research International 25, 333-341.

Godfray,H.C.J, Beddington,J.R., Crute,I.R., Haddad, L., Muir, J.F.,Pretty,J.,Robinson,S.,

Thomas,S.M. and Toulmin,C. 2010. Food Security: the Challenge of Feeding 9 Billion

People. Science 327, 812–18.

Godfray, H. C. J., and Garnett, T. 2014. Food security and sustainable

intensification. Philosophical Transactions of the Royal Society B: Biological Sciences 369,

1639.

http://www.fao.org/fishery/species/3105/en

157

Gómez-Guillén, M.C., Giménez, B., López-Caballero, M.E. and Montero, M.P. 2011.

Functional and bioactive properties of collagen and gelatin from alternative sources: A

review Food Hydrocolloids 25(8),1813-1827

Gudmundsson, M. (2002). Rheological properties of fish gelatins. Journal of Food Science,

67, 2172-2176.

Guerrero, P., Kerry, J.P. and de la Caba, K. 2014. FTIR characterization of protein–

polysaccharide interactions in extruded blends. Carbohydrate Polymers 111, 598–605.

Gunaratne, A. and Hoover, R. 2002. Effect of heat-moisture treatment on the structure and

physicochemical properties of tuber and root starches. Carbohydrate Polymers 49, 425-437.

Guo,W. and Greaser,M.L. 2017. Muscle Structure, Proteins, and Meat Quality. In: New

Aspects of Meat Quality From Genes to Ethics, Pages Purslow,P.P.(Ed) Woodhead

Publishing Series in Food Science, Technology and Nutrition pp 13-31.

Hama,F., Icard-Vernière,C. Guyot, J-P.,Picq, C., Diawara, B. and Mouquet-Rivier,C. 2011.

Changes in micro- and macronutrient composition of pearl millet and white sorghum during

in field versus laboratory decortication. Journal of Cereal Science 54, 425-433.

Hames, B. D., and D. Rickwood. 1990. Gel electrophoresis of proteins: a practical

approach. IRL Oxford Press, England pp 1–91.

Han, Z., Cai, M., Cheng,J-H. and Sun, D-W. 2018. Effects of electric fields and

electromagnetic wave on food protein structure and functionality: A review. Trends in Food

Science and Technology 75, 1–9.

Hara, M.M. 2001. Could marine resources provide a short-term solution to declining fish

supply in SADC inland countries? The case of horse mackerel. Food Policy 26, 11–34.

158

Haug, I. J., Draget, K. I., and Smidsrød, O. (2004). Physical and rheological properties of fish

gelatin compared to mammalian gelatin. Food Hydrocolloids, 18, 203-213.

Hemung, B.-O., Chin.K. B. 2013. Effects of fish sarcoplasmic proteins on the properties of

myofibrillar protein gels mediated by microbial transglutaminase. LWT – Food Science and

Technology 53, 184-190.

Hoover, R., Swamidas, G., Kok,L.S. Vasanthan, T.1996. Composition and physicochemical

properties of starch from pearl millet grains, Food Chemistry 56(4), 355-367.

Howell,N.K.1992. Protein-protein interactions. In: Biochemistry of Food Proteins,

B.J.F.Hudson (Ed) Elsevier Applied Science pp 35-74.

Howell, N.K. and Li Chan, E. 1996. Elucidation of interactions of lysozyme with whey

proteins by Raman Spectroscopy. Int. J. Food Sci. Technol. 31, 439-451.

Howell, N.K., Arteaga, G., Nakai, S. and Li-Chan, E.C.Y. (1999). Raman spectral analysis in

the C-H stretching region of proteins and amino acids for investigation of hydrophobic

interactions. J. Agric. Food Chem. 47, 924-933.

Howell,N.K. 2000. Recent advances in protein interactions. In: Gums and Stabilisers in the

Food Industry, P.A.Williams and G.O.Philipps (Eds), Royal Society of Chemistry pp 317-

327.

Hudson, B.J.F. and Ogunsua, A.O.1974. Lipids of Cassava Tubers (Manihot esculenta

Crantz). J. Sci. Fd Agric. 25, 1503-1508.

Hwang D.C. and Damodaran.S. 1997. Synthesis and Properties of Fish Protein-Based

Hydrogel. JAOCS 1165-1171.

Ignjatovic-Micic,D.,Vancetovic,J.,Trbovic,D.,Dumanovic,Z.,Kostadinovic,M.and

Bozinovic,S. 2014. Grain Nutrient Composition of Maize (Zea mays L.) Drought-Tolerant

Populations. J. Agric. Food Chem. 63, 1251−1260.

159

Jabeen, F. and Chaudhry, A.S. 2011. Chemical compositions and fatty acid profiles of three

freshwater fish species. Food Chemistry 125, 991–996.

Jane,J.,Chen,Y.Y.,Lee,L.F.,McPherson,A.E.,Wong,K.S., Radosavljevic,M. and Kasemsuwan,

T. 1999. Effects of Amylopectin Branch Chain Length and Amylose Content on the

Gelatinization and Pasting Properties of Starch. Cereal Chemistry 76(5), 629–637.

Jardas, I., Šantić, M. and Pallaoro, A. 2004. Diet composition and feeding intensity of horse

mackerel, Trachurus trachurus (Osteichthyes: Carangidae) in the Eastern Adriatic. Marine

Biology 144, 1051-1056.

Jellum, M.D. 1971. Fatty acid composition of corn endosperm and germ oils as influenced by

different extraction procedures. Journal of the American Oil Chemists’ Society 48(7), 355-

357.

Josephson, A.L., Ricker-Gilbert, J.,Florax, R.J.G.M. 2014. How does population density

influence agricultural intensification and productivity? Evidence from Ethiopia. Food Policy

48, 142-152.

Karim, A.A. and Bhat, R. 2009. Fish gelatin: Properties, challenges, and prospects as an

alternative to mammalian gelatins. Food Hydrocolloids 23, 563–576.

Kaur, N., Chugh, V. and Gupta, A.K.2014. Essential fatty acids as functional components of

foods- a review Journal of Food Science and Technology 51(10), 2289–2303

Kristinsson, H.G. and Rasco,B.A. 2000. Fish Protein Hydrolysates: Production, Biochemical,

and Functional Properties. Critical Reviews in Food Science and Nutrition 40 (1), 43-81

Kűçükgülmez, A., Celik, M., Ersoy, B. and Yanar, Y. 2010. Effects of season on proximate

and fatty acid compositions of two Mediterranean fish – the round herring (Etrumeus teres)

and tub gurnard (Chelidonichthys lucernus). International Journal of Food Science and

Technology 45, 1056–1060.

160

Landry, J. and Moureaux, T. 1980. Distribution and Amino Acid Composition of Protein

Groups Located in Different Histological Parts of Maize Grain. J. Agric. Food Chemistry 28,

1187-1 191.

Lange, G-M.2003. The value of Namibia’s commercial fisheries.DEA Research Discussion

Paper Number 55.

https://millenniumindicators.un.org/unsd/envAccounting/ceea/archive/Fish/Namibia_Value_

Commercial_Fisheries.PDF (accessed 06 October 2018).

Larsen, E., Heldbo, J., Jespersen, C. M., and Nielsen, J. (1992). Development of a method

for quality assessment of fish for human consumption based on sensory evaluation.

In H. H. Huss, M. Jakobsen, and J. Liston (Eds.), Quality assurance in the fish

Industry, Elsevier pp. 351-357, Amsterdam.

Lauritzen,L.,Brambilla,P.,Mazzocchi,A.,Harsløf,L.B.S,Ciappolino,V. and Agostoni, C. 2016.

DHA Effects in Brain Development and Function. Nutrients. 8(1), 6.

Lazou, A. and Krokida, M.2010. Structural and textural characterization of corn–lentil

extruded snacks. Journal of Food Engineering 100, 392–408

Li, S., Xu, X. and Zhou.G. 2012. The roles of the actin-myosin interaction and proteolysis in

tenderization during the aging of chicken muscle. Poultry Science 150-160.

Liu, R., Zhao, S., Xiong, S., Qi, C. and Xie, B. 2007. Rheological Properties of fish

actomyosin and pork actomyosin solutions. Journal of Food Engineering 85, 173-179.

Liu,H.Y.,,Li,D. and Guo,S. 2007. Studies on collagen from the skin of channel catfish

(Ictalurus punctaus) Food Chemistry 101, 621–625.

https://millenniumindicators.un.org/unsd/envAccounting/ceea/archive/Fish/Namibia_Value_Commercial_Fisheries.PDF

https://millenniumindicators.un.org/unsd/envAccounting/ceea/archive/Fish/Namibia_Value_Commercial_Fisheries.PDF

161

Lupin,H.M., Parin,M.A. and Zugarramurdi,A. 2010. HACCP economics in fish processing

plants. Food Control 21,1143-1149.

Lyberg,A.M., Fasoli,E. and Adlercreutz, P. 2005. Monitoring the oxidation of

docosahexaenoic acid in lipids. Lipids 40(9), 969-79.

Martone, C. B. Busconi, L. Folco, E. J., Trucco,R. E. and Sanchez.,J. J. 1986. A Simplified

Myosin Preparation from Marine Fish Species. Journal of Food Science 1554-1555.

Meng, X., Threinen, D., Hansen, M. and Driedger, D. 2010. Effects of extrusion conditions

on system parameters and physical properties of a chickpea flour-based snack Food Research

International 43, 650–658.

Metillo, E. and Aspiras-Eya, A.A.2014. Fatty Acids in Six Small Pelagic Fish Species and

Their Crustacean Prey from the Mindanao Sea, Southern Philippines Tropical Life Sciences

Research 25(1), 105–115

Montagnac, J.A., Davis, C.R. and Tanumihardjo, S.A. 2009. Nutritional Value of Cassava for

Use as a Staple Food and Recent Advances for Improvement. Comprehensive Reviews in

Food Science and Food Safety 8, 181-194.

Moraru, C.I. and Kokini, J.L.2003. Nucleation and Expansion During Extrusion and

Microwave Heating of Cereal Foods. Comprehensive Reviews in Food Science and Food

Safety 2, 148-165.

Moreno, H.M., Jacq, C., Montero, M.P., Gómez-Guillén, M.C., Borderías, A.J. and Mørkøre,

T. 2016. Effect of selective breeding on collagen properties of Atlantic salmon (Salmo salar

L.). Food Chemistry 190, 856-863.

Morgan, A.R.1984. Controlling food plant quality costs. Food Tech., EE.UU. April:111-112.

162

Mukumangeni, R. and Wiium, W.2006. The Namibian Horse Mackerel Fishery: Juveniles or

Adults. Final Project Report. Icelandic International Development Agency (ICEIDA).The

United Nations University Fish Training Programme.

Muyonga, J.H., Cole, C.G.B. and Duodu, K.G. 2004. Characterisation of acid soluble

collagen from skins of young and adult Nile perch (Lates niloticus). Food Chemistry 85, 81–

89.

Nagaia, T., Arakib, Y and Suzukia.N. 2002. Collagen of the skin of ocellate puffer fish

(Takifugu rubripes). Food Chemistry 78, 173–177.

Nambiar,V.S., Dhaduk, J.J., Sareen, N., Shahu, T. and Desai, R. 2011.Potential Functional

Implications of Pearl millet (Pennisetum glaucum) in Health and Disease Journal of Applied

Pharmaceutical Science 1(10), 62-67

Nascimento,A.C., Mota,C., Coelho, I., Gueifão,S., Santos,M., Matos,A.S.,Gimenez,A.,

Lobo,M., Samman,N. and Castanheira,I. 2014. Characterisation of nutrient profile of quinoa

(Chenopodium quinoa), amaranth (Amaranthus caudatus), and purple corn (Zea mays L.)

consumed in the North of Argentina: Proximates, minerals and trace elements. Food

Chemistry 148, 420–426.

Nehete, J.Y., Bhambar, R.S., Narkhede, M.R. and Gawali, S.R. 2013. Natural proteins:

Sources, isolation, characterization and applications. Pharmacogn Rev 7(14), 107–116.

Nishimoto, M., Sakamoto, R., Mizuta, S. and Yoshinaka, R. 2005. Identification and

characterization of molecular species of collagen in ordinary muscle and skin of the Japanese

flounder Paralichthys olivaceus Food Chemistry 90, 151–156.

Ngudi, D.D., Kuo, Y-H and Lambein, F. 2002. Food Safety and Amino Acid Balance in

Processed Cassava “Cossettes”.J. Agric. Food Chem 50(10), 3042–3049.

163

Ninan, G., Joseph, J. and Zynudheen Abubacker Aliyamveettil, Z.A. 2014. A comparative

study on the physical, chemical and functional properties of carp skin and mammalian

gelatins. Journal of Food Science and Technology 51(9), 2085–2091.

Omobuwajo, T.O. 2003. Compositional characteristics and sensory quality of biscuits, prawn

crackers and fried chips produced from beadfruit. Innovative Food Science and Emerging

Technologies 4 , 219–225.

Onwulata, C.I. and Qi, P.X. 2006. Food Proteins: Interactions and Functionality. In book:

Encyclopedia of Molecular Cell Biology and Molecular Medicine.

Orban, E. Di Lena, G. Nevigato, T. Masci, M. Casinii, I. and Caproni, R. 2011. Journal of

Food Composition and Analysis 24, 1110–1116.

Orsavova, J., Misurcova, L., Ambrozova, J.V., Vicha, R. 3 and Mlcek, J. 2015. Fatty Acids

Composition of Vegetable Oils and Its Contribution to Dietary Energy Intake and

Dependence of Cardiovascular Mortality on Dietary Intake of Fatty Acids. Int. J. Mol. Sci.

16, 12871-12890

Osako, K., Kuwahara, K, Saito, H., Hossain, M.A. and Nozaki, Y. 2003. Effect of Starvation

on Lipid Metabolism and Stability of DHA Content of Lipids in Horse Mackerel (Trachurus

japonicus) Tissues. Lipids 38(12), 1264-1267.

Osman, M.A. 2011. Effect of traditional fermentation process on the nutrient and antinutrient

contents of pearl millet during preparation of Lohoh. Journal of the Saudi Society of

Agricultural Sciences 10, 1–6.

Pansawat, N., Jangchud, K., Jangchud, N., Wuttijumnong, P., Saalia, F.K., Eitenmiller, R.R.

and Phillips, R.D. 2008. Effects of extrusion conditions on secondary extrusion variables and

164

physical properties of fish, rice-based snacks. LWT - Food Science and Technology 41(4),

632-641.

Park,J-M., Kwon,S-H., Han,Y-M.,Hahm,K-B. and Kim,E-H. 2013. Omega-3 Polyunsaturated

Fatty Acids as Potential Chemopreventive Agent for Gastrointestinal Cancer. J Cancer Prev.

18(3), 201–208.

Patil, R. T., Berrios, J. D. J., Tang, J., and Swanson, B. G. (2007). Evaluation of the methods

for expansion properties of legume extrudates. Applied Engineering in Agriculture 23(6),

777–783.

Pérez-Alvarez, J., Fernández-López, J. 2007. Chemical and Biochemical Aspects of Colour

in Muscle Foods. In: L.M.L Nollet,T. Boylston, F.Chen,P.C. Coggins,M.B. Gloria,G. Hyldig,

C.R.Kerth,L.H.McKee and Y.H.Hui,(Eds).L.M.L.2007.Handbook of Meat, Poultry and

Seafood Quality. Blackwell Publishing. Pp.1-24.

Petrofsky,K.E. and Hoseney,R.C. 1995.Rheological Properties of Dough made with Starch

and Gluten from Several Cereal Sources Cereal Chemistry 72 (1) 53-58.

Prasanna B. M., Vasal S. K., Kassahun, B and N. N. Singh. 2001.Quality protein maize.

Current Science 81 (10) 1308-1319.

Prošková, J. 2014. Description of protein secondary structure using dual quaternions. Journal

of Molecular Structure 1076, 89–93.

Qamar, S., Aslam, M. and Javed, M.A.2016. Determination of Proximate Chemical

Composition and Detection of Inorganic Nutrients in Maize (Zea mays L.) Materials Today:

Proceedings 3(2), 715-718

165

Ramalhosa, M.J. Paíga, P. Morais, S. Alves, M.R. Delerue-Matos, C. and Oliveira, M.B.P.P.

2012. Lipid content of frozen fish: Comparison of different extraction methods and

variability during freezing storage. Food Chemistry 131, 328–336.

Ramírez, J.A., Martin-Polo, M. O. and Bandman, E. 2000. Fish Myosin Aggregation as

Affected by Freezing and Initial Physical State. Journal of Food Science: Food Chemistry

and Toxicology 556-560.

Riaz,M.N. 2013. Food Extruders. In: Handbook of Farm, Dairy and Food Machinery

Engineering (Second Edition), M.Kutz.(Ed), Academic Press. Pages 427-440.

Rayan, A.M. and Abbott, L.C. 2015. Compositional analysis of genetically modified corn

events (NK603, MON88017 x MON810 and MON89034 x MON88017) compared to

conventional corn. Food Chemistry 176, 99–105.

Rolf Kohl, former Dr.rer.nat. (equivalent Ph.D.), Dipl. Biol. at Nutrition,

https://www.quora.com/Does-seaweed-contain-collagen (accessed on October 3, 2018).

Rooney, L.W. 1978. Sorghum and Pearl Millet Lipids. Cereal Chemistry 55(5), 584-590.

Saaed, S.1998. Phd Thesis. Lipid Oxidation mechanisms and lipid-protein interactions in

frozen mackerel (Scomber scombrus).University of Surrey. Guildford. United Kingdom.

Saeed, S and N. K. Howell. 1999. High-performance liquid chromatography and

spectroscopic studies on fish oil oxidation products extracted from frozen Atlantic

mackerel. J. Am. Oil Chem. Soc. 76, 391–397.

Sakko, A. L. 1998. The influence of the Benguela upwelling system on Namibia's marine

biodiversity. Biodiversity and Conservation 7, 419-433.

Sandholm, L. 1987. Reducing quality costs to improve export earnings. International Trade

Forum, Oct.-Dec, 26-31.

166

Sarbon,N.M.,Badii,F. and Howell,N.K. 2015. The effect of chicken skin gelatin and whey

protein interactions on rheological and thermal properties. Food Hydrocolloids 45, 83-92.

Sarkadei, S. 2005. Mechanisms of Lipid Oxidation and safety assessment in underutilised fish

species. Ph.D. Thesis. University of Surrey. Guildford. United Kingdom.

Sawaya, W.N., Khalil, J.K. and Safi, W.J.1984. Nutritional quality of pearl millet flour and

bread. Plant Foods for Human Nutrition 34(2), 117–125

Schönfeldt, H.C. and Hall, N.G. 2011.Determining iron bio-availability with a constant heme

iron value. Journal of Food Composition and Analysis 24(4-5), 738-740.

Sen, D.P. Chemical composition and their technological significance. 2005. In: Advances in

Fish Processing Technology.Sen., D.P.Allied Publishers Pvt.Ltd.New Delhi.

Serna-Saldivar, S.O. and Rooney, L.W. 2015. Industrial Production of Maize Tortillas and

Snacks. In: Tortillas Wheat Flour and Corn Products. L.W. Rooney and S.O. Serna-Saldivar

(Eds). Woodhead Publishing and AACC International Press pp 247-281.

Shahidi, F. 1994. Seafood proteins and preparation of protein concentrates. In: Seafoods:

Chemistry, Processing Technology and Quality pp 3-9.

Shahidi, F. and Botta, J.R.2012. In: Seafoods: Chemistry, Processing Technology and Quality

Springer Science and Business Media p 4.

Shewry.P.R. and Halford, N.G. 2002.Cereal seed storage proteins: structures, properties and

role in grain utilization. J. Exp. Bot. 53(370), 947-958.

Sikorski, Z. 2012. Seafood Proteins Springer Science and Business Media, p 15.

Silva, H. A., Mendes, R., Nunes, M. L. and Empis, J. 2006. Protein changes after irradiation

and ice storage of horse mackerel (Trachurus trachurus). Eur Food Res Technol 224: 83–90.

Simpson, B.K. 2012. Food Biochemistry and Food Processing. John Wiley and Sons p 319.

167

Slavin,J.L, Jacobs,D.and Marquart,L. 2001. Grain Processing and Nutrition.Crit.Rev.Biotech.

21 49-46.

Steiner-Asiedu, M., Lied, E., Nilsen, R. and Julshamn, K. 1993. The Nutritive Value of Sun-

Dried Pelagic Fish from the Rift Valley in Africa. J Sci Food Agric 63, 439-443.

Sterling R. 1985. Relationship between manufacturing and quality costs. Food Tech.,

EE.UU., September: 54-55.

Stojceska, V., Ainsworth, P., Plunkett, A. and İbanoǧul, Ş. 2009.The effect of extrusion

cooking using different water feed rates on the quality of ready-to-eat snacks made from food

by-products. Food Chemistry 114, 226–232.

Sun, X. D. and Holley, R. A. 2010. Factors Influencing Gel Formation by Myofibrillar

Proteins in Muscle Foods. Comprehensive Reviews in Food Science and Food Safety 33-46.

Tacer-Caba, Z., Dilara Nilufer-Erdil, D., and Ai, Y. 2015. Chemical Composition of Cereals

and Their Products.In: Handbook of Food Chemistry. P.C.K. Cheung and B.M. Mehta pp 1-

23.

Tan, Y., Gao, H., Chang, S.K.C., Bechtel, P.J. and Mahmoud, B.S.M.2019. Comparative

studies on the yield and characteristics of myofibrillar proteins from catfish heads and frames

extracted by two methods for making surimi-like protein gel products. Food Chemistry 30

(272), 133-140

Tang,M., He, X., Luo,Y., Ma,L., Tang,X. and Huang,K. 2013. Nutritional assessment of

transgenic lysine‐rich maize compared with conventional quality protein maize. Journal of

the Science in Food and Agriculture 93(5), 1049-105

Tester, R.F. and Morrison, W.R. 1990. Swelling and Gelatinization of Cereal Starches. II.

Waxy Rice Starches. Cereal Chemistry 67(6), 558-563.

168

Tomlinson, I. 2013. Doubling food production to feed the 9 billion: A critical perspective on

a key discourse of food security in the UK.Journal of Rural Studies 29, 81-90.

Toppe, J., Albrektsen, S., Hope, B. and Aksnes, A. 2007. Chemical composition, mineral

content and amino acid and lipid profiles in bones from various fish species. Comp Biochem

Physiol B Biochem Mol Biol. 146(3), 395-401.

Tornberg,E. 2005. Effects of heat on meat proteins - Implication on structure and quality of

meat products. Meat Science 70(3), 493-508

Usydus, Z.,Szlinder-Richert,J.,Adamczyk,M. 2009. Protein quality and amino acid profiles of

fish products available in Poland. Food Chemistry 112(1), 139-145.

van Ittersum,M.K., van Bussel,L.G.J, Wolf,J. P.,Grassini,P., van Wart,J., Guilpart,N.

Claessens, L., de Groot, H., Wiebe,K., Mason-D’Croz,D., Yang,H.,Boogaard,H.,van

Oort,P.A.J.,van Loon,M.P., Kazuki Saito,K., Adimo,O., Adjei-Nsiah,S., Agali,A.,Bala,A.,

Chikowo,R., Kaizzi,K.,Kouressy,M., Joachim H. J. R. Makoi,J.H.J.R., Ouattara,K.,

Tesfaye,K. and Cassman,K.G. 2016. Can sub-Saharan Africa feed itself? Proc Natl Acad Sci

U S A. 113(52), 14964–14969.

Vazquez, J.F. and Gómez-Aldapa, C.A. 2001. Chemical and Physicochemical Properties of

Maize Starch After Industrial Nixtamalization. Cereal Chemistry 78( 5), 543 -550.

Venugopal, V. and Shahidi, F. 1996. Structure and composition of fish muscle. Food Reviews

International 12(2), 175-197.

Venugopal, V. 2008. Marine Products for Healthcare: Functional and Bioactive

Nutraceutical. CRC Press p 53.

Vieira, S., Franco, A.R., Fernandes, E.M., Amorim, S., Ferreira, H., Pires, R.A., Reis, R.L.,

Martins, A. and Neves, N. 2018. Fish sarcoplasmic proteins as a high value marine material

for wound dressing applications. Colloids Surf B Biointerfaces167, 310-317.

169

Wang,S., Zhang,X., Wang, S. and Copeland, L. 2016. Changes of multi-scale structure

during mimicked DSC heating reveal the nature of starch gelatinization. Sci Rep. 6, 28271.

Waterschoot, J., Gomand, S.V., Fierens, E. and Delcour, J.A. 2015. Production, structure,

physicochemical and functional properties of maize, cassava, wheat, potato and rice starches.

Starch/Stärke 67, 14–29.

Weber, E.J. 1978. Corn Lipids. Cereal Chemistry 55(5), 572-584.

Weder, J.K., Sohns, S. 1983.Model studies on the heating of food proteins. Amino acid

composition of lysozyme, ribonuclease and insulin after dry heating. Z Lebensm Unters

Forsch. 176(6), 421-5.

WFO.2017. Namibia Country Strategic Plan (2017–2022). Executive Board Annual Session

Rome, 12–16 June 2017.

Willett, J.L., Millard, M.M. and Jasberg, B.K. 1997. Extrusion of waxy maize starch: melt

rheology and molecular weight degradation of amylopectin Polymer 38(24), 5983-5989.

Wilson, L.M. Whitt, S.R., Ibáñez, A.M., Rocheford, T.R., Goodman, M.M. and Buckler, E.S.

2004. Dissection of Maize Kernel Composition and Starch Production by Candidate Gene

Association. The Plant Cell 16, 2719–2733.

Xiong,Y.L. 2018. Muscle Proteins. In: Proteins in Food Processing (Second Edition),

R.Y.Yada (Ed), Woodhead Publishing Series in Food Science, Technology and Nutrition pp

127-148.

Yadav,D.N., Chhikara,N., Anand, T. and Sharma,M.2014. Rheological quality of pearl millet

porridge as affected by grits size Journal of Food Science and Technology 51(9), 2169-2175

170

Yankova, M. H., Raykov, V. S. and Frateva. P. B. 2008. Diet Composition of Horse

Mackerel, Trachurus mediterraneus ponticus Aleev, 1956 (Osteichthyes: Carangidae) in the

Bulgarian Black Sea Waters. Turkish Journal of Fisheries and Aquatic Sciences 8: 321-327

Yildiz, F. 2009. In: Advances in Food Biochemistry, F.Yildiz (Ed), CRC Press p 77.

Yongfeng, A. and Jay‐lin, J. 2014. Gelatinization and rheological properties of starch.Starch

67(3-4), 213-224.

Yüksektepe,F.Ű, Yılmaz,Ő, and Türkay, M. 2008. Prediction of secondary structures of

proteins using a two-stage method. Computers and Chemical Engineering 32, 78–88.

Zeng, J, Gao, H. and Li, G. 2014. Functional properties of wheat starch with different particle

size distribution. J Sci Food Agric. 94, 57-62.

Zugarramurdi, A., Parin, M.A., Gadaleta, L., Carizzo,G. and Lupin, H.M. 2004.The effect of

improving raw material quality on product quality and operating costs: a comparative study

for lean and fatty fish. Food Control 15, 503-509.

Zugarramurdi, A., Parin, M.A., Gadaleta, L. and Lupin, H.M. 2007. A quality cost model for

food processing plants. Journal of Food Engineering 83, 414-421.

Zugarramurdi, A. 2013. 18 Month report Securefish Project, presented June 2013.

171

9. APPENDICES

Appendix 1.1 List of categories and elements of quality costs (Zugarramurdi et al., 2007)

Controllable costs Resulting costs

Prevention Costs Internal Failure Costs

P1. Design, development and implementation

of a quality assurance plan

F1. Scraps, reprocessing or spoilage

P2. Quality training programs for suppliers

and production personnel

F2. Low labour productivity and low process

yield

P3. Hygiene and sanitation of the plant F3. Inefficient usage of plant capacity

P4. Preventive maintenance and additional

supervision

Appraisal Costs

A1. Receipt and control of incoming material

A2. Sampling and laboratory analysis

A3. In-process inspection

172

Appendix 1.2. Quality, market and production parameters (Zugarramurdi et al., 2007)

Quality

parameters:

Qm = raw material quality (normalized parameter)

Qp = product quality (normalized parameter)

YQm* = yield for optimum quality level Qm* (kg product/kg raw

material)

XQm* = productivity for optimum quality level Qm* (kg product/h-

worker)

Nscp = number of sanitation control points

Nccp = number of critical control points

Market

parameters:

PQm = purchase price of raw material (US$/kg)

PQp = selling price for quality level Qp (US$/kg)

PQp* = selling price for optimum quality level Qp* (US$/kg)

Production

parameters:

d = number of working days per year

K = daily production (kg finished product/day)

Ri = raw material sampling inspection rate (kg raw material/h)

S = average labour rate for trained workers (US$/h)

IF = total fixed investment (US$)

173

XQm = productivity for quality level Qm (kg product/h-worker)

YQm= yield for quality level Qm (kg product/kg raw material)

L = total labour cost (US$/kg product)

Appendix 1.3. Relationships used to estimate quality costs components (Zugarramurdi et al.,

2007)

Type of cost Relationship Equation

Number

Controllable costs sub-

model

Prevention costs

P1. Design, development

and implementation of a

quality assurance plan

CP1 = ΩP1 IF / d K (1)

P2. Quality training

programs for suppliers and

production personnel

CP2 = ΩP2 L Qp (2)

P3. Hygiene and sanitation

of the plant

CP3 = (ΩP3 L [1 + 0.01 Nscp ] Qp) / z (3)

174

P4. Preventive maintenance

and additional supervision

CP4 = [ΩP4 IF / (d K) + ΩP5 L ] Qp (4)

Appraisal Costs

A1. Receipt and control of

incoming material

CA1 = ( ΩA1 S / YQm Ri ) Qp (5)

A2. Sampling and

laboratory analysis

CA2 = ΩA2 L Qp (6)

A3. In-process inspection CA3= ΩA3 Nccp L Qp (7)

Internal Failure Costs

F1. Scraps, reprocessing or

spoilage

CF1 = (PQp* - PQp ) (1 - Qp)2 (8)

F2. Low labour

productivity and low

process yield

CF2 = S (1/ XQm - 1 / XQm* ) + PQm (1/

YQm - 1/ YQm*)

(9)

F3. Inefficient usage of

plant capacity

CF3 = PQp (YQm* - YQm ) / YQm* (10)

physico-biochemical properties of extruded cape horse ...epubs.surrey.ac.uk › 852198 › 1 ›...

Documents